Auricle Publishing
Number 6 and 7 ©2001- Revised 2008
4 Crisis in the Industry Speakers Part 2 Introduction 8 These speakers sound better than conventional “let’s Time and Phase Audio Perfectionist Journals #6 and #7 have been put some drivers-in-a-box” designs yet they cost no Not Just a Craze combined for improved continuity. It took several more. These speakers can provide more long-term months to gather this data and write the more than satisfaction than exotic “like nothing you’ve ever 20,000 words contained herein. Rather than split this heard before” designs and often cost far less. 10 information into two consecutive Journals, I decided Dunlavy to present everything together in a double issue. Are these speakers perfect? Hardly. Don’t all speak- ers with accurate response sound the same? No, and We’ll continue the discussion of loudspeaker accu- it would certainly be a dull world if they did. 15 racy and I’ll restate some of my views on the impor- Dunlavy Graphs tance of accurately replicating the information on Each of the speaker brands represented here has a the recording. We’ll talk about what high quality sonic signature that is shared by all models in that loudspeakers should cost and about the cost-versus- manufacturer’s line, but the sonic characteristics of 19 performance crisis in the industry. A guest writer will each brand are slightly different. Thiel share some of his personal experiences with Journal readers. These speakers have the capability of accurately reproducing the input signal, as demonstrated by the 25 Previously, we examined some speaker systems objective measurements, but no model described Thiel Graphs which are demonstrably inaccurate and we discussed here is completely free of coloration or compromise. the reasons why many speakers fail to accurately reproduce the input signal. In this Journal we’ll We have discussed how speakers can go wrong but 29 examine some speaker systems which are demon- measurements can’t tell us everything about speaker Vandersteen strably accurate, according to all the accepted objec- sound. Objective measurements can only tell us tive standards. which products are capable of accurate response and which ones can’t possibly reproduce music correctly. 36 These speakers have flat frequency response within Vandersteen narrow limits so they don’t emphasize some frequen- Measurements can clearly Graphs cies while de-emphasizing others. These speakers are time- and phase-correct so they won’t reproduce show when the design some harmonic overtones out-of-phase with the fun- process has failed. 40 damental tones, smearing transient sounds over time. Journey to Measurements can clearly show when the design Enlightenment These speakers don’t sing along after the song has ended because they are carefully designed and con- process has failed by demonstrating that a given structed to minimize electrical and mechanical energy speaker has gross deviations from flat frequency storage and delayed reflections. response or is incoherent in the time domain. While 45 poor measurements can guarantee that a product will Conclusion
www.audioperfectionist.com Introduction be unsatisfying in the long term, good measurements can only increased by making the colors in the photo a little more “color- guarantee the potential for good sound. ful.” Others consider the act of viewing a photograph to be an entirely different experience from the act of viewing a sunset. Our ears can still hear a lot more than our instruments can They feel that there should be no limitations at all on how far measure and, while much of speaker design is based in sci- the photo can deviate from a realistic presentation so long as ence, there is still some art involved. it produces the desired emotional response. Similar arguments occur in audio. Each of the three designers profiled in this issue has applied different engineering techniques while attempting to solve the Some feel that the job of an audio system is to accurately speaker design problems delineated in Journal #5. These reproduce the recorded information. Those of us who feel this problems cause speakers to deviate from accurate reproduction way think that the artists who make the recording should decide of the input signal. All speaker designers must confront these how that recording will sound, not the playback system. We same issues but some have been far more successful in over- assume that the artists and engineers who make the recording coming these challenges. In the articles that follow I’ll profile are attempting to bring the live performance into our homes so three of the designers who have been the most successful, in that we can experience at home what they experienced during my opinion, and describe their individual approaches to the sci- the actual event. ence and art of loudspeaker design. In this double issue we’ll discuss design philosophies and in the next Journal we’ll talk Some feel that the job more about the sound of these products. of an audio system is to Before we begin, let’s revisit the subject of accuracy and the philosophical arguments for its importance. While I believe that accurately reproduce accurate reproduction of the recorded signal is the ultimate goal of a high fidelity audio system, this is an arbitrary position that the recorded information. is not universally accepted. Those of us who accept this philosophy believe that the audio system should resolve as much information from the record- In The Beginning ing as possible and should not impose its characteristics on the sound we hear during playback. We think that recordings In the premier issue I made some statements about the philos- should be presented “as is” with no embellishment. There is ophy I would follow in writing the Audio Perfectionist Journal another point of view. and I presented an article titled Accuracy that included argu- ments supporting my position on that subject. Listening to music at home and attending a live performance are two very different experiences, in the opinion of some. This I believe that a high fidelity audio system should attempt to viewpoint holds that the two activities have little in common and reproduce the recorded signal as faithfully and accurately as should be considered separately. After all, the musicians aren’t possible. Others disagree—even though they might not admit really there in your living room helping to get the audience that fact aloud or in print. Many believe that the experience of involved in the music. listening is of primary importance, and how that experience is produced is of little significance. This, too, is a valid position. Some think that a home audio system needs to present sound Here is an analogy. that is a little more spectacular than what you’d hear at the live event in order to provide the listener with similar emotional Viewing a photograph of a beautiful sunset will seldom produce gratification. These folks feel that “goosing up” the response the emotional response of seeing the real thing. While many of a home audio system is a perfectly acceptable thing to do. photographers try to capture the actual scene as faithfully as The arrival of home theater has increased the numbers in this possible, others strive primarily to create an emotional effect on camp. the viewer. Some feel that the impact of the photograph can be
2 www.audioperfectionist.com Introduction
Movie sound doesn’t even try to mimic real sound. Everything Journal #5 presented measurement graphs of speakers which is exaggerated for effect. That’s why most movie sounds are are demonstrably inaccurate. This issue will present graphs called “effects.” Real people don’t make that much noise when from speakers which are demonstrably accurate. In Journal #8 they walk or when they eat. Romantic moments in real life are I’ll show how some measurements can actually be misleading seldom accompanied by theme music. Real fists seldom make and we’ll discuss the subjective aspects of speaker sound. that much noise when they strike real flesh. There is no “real” standard for sound effects, which were artificially created to There is probably not a more controversial subject than the start with. THX has added to the confusion about the goal of a argument about whether the time and phase characteristics of home audio system. Should the system accurately reproduce a loudspeaker are audible. Many “authorities” with PhDs after the recorded sound track or should the system alter the record- their names claim to have proven that average listeners cannot ed information in an attempt to make the living room sound like hear the effects of phase shift in loudspeakers. While there is a movie theater? no doubt in my mind that these characteristics are audible and important this is still the subject of intense debate. Journal #5 presented some of my views on the subject. Journals #6/7 & Some think that a home audio #8 will present some more. I advise you to read all you can about the subject and then go and listen to make your own system needs to present sound determination. that is a little more spectacular In This Issue than what you’d hear at the live The article titled Crisis in the Industry introduces this issue and states some of my motives for writing it and launching the event... Audio Perfectionist Journal in the first place. Many so-called “high-end” speakers cost too much and perform poorly. This I have found that honesty is the best policy, in audio as in life. article presents some of the reasons for this situation. I think that an audio system should accurately reproduce the recorded information, whatever that may be. I think that this The article titled Time and Phase, Not Just a Craze presents philosophy provides the best path to long-term satisfaction. my views on the importance of time-domain performance in This is an arbitrary choice based on my personal experience loudspeaker design. This is a very controversial subject and it and others can effectively argue for other points of view. is not the primary thrust of this Journal. If you don’t feel that this aspect of performance is audible, please don’t abandon the entire issue. There is plenty of other valuable information In The Last Issue presented here. In Journal #5 we discussed some of the problems that design- ers face when trying to create an accurate loudspeaker system. There are three major manufacturers making time- and phase- The fact that many designers don’t even make an attempt at accurate speakers and there may be many smaller ones of accurate reproduction was mentioned and the reasons why this which I am not aware. In this Journal I’m going to write about is true were presented. each of the three with which I am familiar, in alphabetical order. I’ll discuss the approach that each designer has taken to over- In this issue we’ll see that some manufacturers do strive for the come the speaker design problems delineated in Journal #5. most accurate reproduction possible and some designers have been very successful in overcoming the inherent problems of This is not meant to be a sales pitch for these brands. My goal accurate electromechanical transduction—changing an electri- is to enlighten the reader about the technology that is available cal signal to an acoustical signal while preserving the amplitude and about what it costs. In my opinion, most people are paying and phase characteristics of that signal. far too much for speakers and getting far less for their money than they should in terms of performance.
www.audioperfectionist.com 3 Crisis in the Industry
This issue contains the first article contributed by someone The Crisis in the Industry other than me. Shane Buettner is Equipment Review Editor of by Richard Hardesty Widescreen Review magazine. He has had the opportunity to listen to virtually all the major and many of the minor speaker The ongoing crisis in the high-end audio industry was one of brands at trade shows, in WSR’s laboratories, and at home my original motivations to launch the Audio Perfectionist during equipment reviews. He has also spent much time in my Journal. It appeared to me that charlatans and snake oil sales- home listening to my personal systems. Shane has contributed men had overrun the seekers of the sonic truth, converting a the article titled Journey to Enlightenment that appears in this once legitimate and vital high-end audio community into an Journal. It’s a story of personal discovery that describes one industry promoting poorly engineered and often ridiculously man’s path toward musical—and home theater—nirvana. overpriced components. I wanted to speak out about this situ- ation and to offer a voice of reason to those who simply want good sound along with good value for their money. Coming Up In Journal #5 we discussed some objective measurements of I believe that knowledge can be a powerful tool in the hands of loudspeaker performance. Measurement graphs from speakers consumers. The more you know about how things are made, which are demonstrably inaccurate were presented along with and about how they work, the less likely you are to be confused suggestions about how to interpret these graphs. Journal #5 by a product review in a magazine; if you understand the prod- provided a list of problems which all speaker designers must ucts that are being discussed and how these products should face. In Journal #6/7 various solutions to these problems are be used you are less likely to be swayed by a skillful sales- discussed and the unique approaches taken by three of the man’s polished pitch. industry’s top engineers are described. Measurement graphs from speakers designed by these engineers are presented. My goal is to provide you with objective, useful information and thoughtful opinions about home audio components and sys- In Journal #8 we’ll continue to examine the subjective qualities tems. It is impossible to do this from any perspective but my of loudspeakers. We’ll talk about how good measurements can own. My ears are attached to my head, after all. sometimes be achieved at the expense of sound quality. While I certainly don’t know everything, I have learned a lot While good measurements can reliably predict which products about our subject of interest over the years. My opinions are will provide unsatisfying performance in the long term, mea- based on my experiences and I can share those with you. I can surements can actually lead you astray in some circumstances. write about the things that I have experienced and the opinions I’ll write more about the sound of the speakers discussed in this I’ve formed, and mention opposing viewpoints that you can issue and I’ll review some specific models of loudspeakers and research elsewhere. You may come to different conclusions in comment on their use for stereo and home theater applications. your own search for sonic satisfaction and that’s part of the fun.
The information in Journals #5 & #6/7 is specifically about In that spirit, I present this issue of the Audio Perfectionist stereo speakers but the same basic data is applicable to speak- Journal. I hope that you will find it interesting and that it will ers used in a home theater system, although some additional provoke you to read other viewpoints and to go out and listen. factors must be considered when speakers are used for center channel and surround channel applications. We’ll discuss these special surround sound applications in Journal #8 and beyond. The Early Years APJ In the beginning, discerning music lovers could buy audio products with acceptable performance at the local chain store, or they could seek out specialty audio retailers where they could often get better sound for the same money. State-of- the-art products were offered to those willing to spend more. “High-end” stores were run, for the most part, by audiophiles
4 www.audioperfectionist.com Crisis in the Industry dedicated to helping music lovers achieve greater satisfaction from their home audio systems. We sold Yamaha, Technics from their home audio systems. They sold products made by and Advent, but we also sold all the top electrostatic speakers specialty manufacturing firms with the same motives. including Dayton-Wright, Sound Lab, Quad, Acoustat, Beverage and Martin-Logan. We sold all the top planar-magnetic speak- Better sound for the money was the rallying cry. Of course very ers including Magneplanar, Eminent Technology and Audire, expensive products, by the standards of the day, were available as well as some true ribbons like Sequerra. We sold all the top there as well but nobody bought them on faith. If it cost more, dynamic speakers like Vandersteen, Thiel, Dahlquist, B&W, the dealer usually had to prove that it sounded better. Things KEF, Rogers, Snell, Braun, M&K, Linn and many others. are different now. We sold all the top vacuum tube electronic lines like Audio Today, it is assumed that a high price tag guarantees high Research, Berning, Audible Illusions, CAL, Theta (Moffat), performance, and the higher the price the better the product is and Luxman. We sold all the top solid-state electronic lines presumed to be. Dealers often feature just a single product at like Mark Levinson, Threshold, GAS, Naim, Linn, Audire, CAL, a given price point and comparisons are seldom allowed and Theta (Sinclair), Proceed, Nakamichi, NAD, Rotel, PSE, PS almost never encouraged. A high price is equated with high Audio, Tandberg, Accuphase, and many, many more. performance but the facts tell a different story. Our goal was to provide the best sound for the money at all price points. That was a common goal throughout the industry Today, it is assumed that in the early days. Value was always a consideration.
a high price tag guarantees Many of the things we sold were expensive but the retail price was always directly related to the cost of manufacturing the high performance. product and we could explain why expensive components cost so much. We would disassemble Mark Levinson amplifiers and Price tags are often completely unrelated to the cost of manu- preamplifiers in front of customers to show a potential buyer facturing the product. Many high-end audio products seem to why these components cost more. We could demonstrate that be made primarily to fleece the wealthy and cheat the unin- our more costly products delivered audibly superior sound by formed, and this drives away music lovers who might otherwise direct comparison. This is almost unheard of today. Value and enter the high-end market. What started with a group of sincere common sense are rare commodities in our modern world. audiophiles trying to make legitimate advances in the audio art has begun to look like a big con game. Common sense suggests that $10,000 speaker cables can- not be justified on any basis related to manufacturing cost, but Transparent Marketing just introduced a cable that sells for The Mass-Market and High-End Audio $23,500 for an 8-foot pair (as reported in Stereophile)! That Over the years, the sound from mass-market electronic prod- report said that it takes two men, two 8-hour work days to make ucts has been steadily improving and the prices for these prod- a set of these cables. Imagine that. Well, two highly skilled men ucts have actually decreased when adjusted for inflation. At the worked for a week-and-a-half remodeling my bathroom and I same time, the price tags on high-end audio components have can assure you that it didn’t cost $23,500 to complete that job! continued to soar ever upward. Many components claiming Who do they think they’re kidding? high-end status offer questionable performance benefits when compared to well-engineered offerings selling for far less. It Speaker cables aren’t the only components with exorbitant wasn’t always that way. profit margins. While the prices (in real dollars) of most mass- market audio components have actually decreased in recent I was there at the genesis of high-end audio. My former partner, years, the prices for high-end audio components have reached Curtis Havens, and I opened one of the first stores catering to levels that would have been unimaginable a decade ago. discerning music lovers who were seeking better performance Whether these incredibly expensive new components actually
www.audioperfectionist.com 5 Crisis in the Industry perform better than those from the previous decade is a matter is about sound quality and how to get it. The fact is that you for debate. Some probably do, but many certainly do not. can get better sound quality for much less. Notice that I didn’t say equivalent sound quality—I said better sound quality. [See Electronic component parts are better today and cost less. Vandersteen 5a review in Journal #12.] Engineers have learned a few things over the years and a thou- sand dollars will buy a much better amplifier today than what No dynamic loudspeaker with a couple hundred dollar’s worth the same money would have purchased a decade ago. While of component parts inside should cost $20,000, even if it is it will definitely cost several thousands of dollars to purchase a properly engineered and meticulously built. Electrostatic and really good solid-state power amplifier, spending $20,000 may planar-magnetic speakers often have substantially lower manu- or may not buy audibly superior performance. (See my review facturing costs than dynamic systems and should sell for less, of the Theta Dreadnaught in Journal #4 for an example.) in most cases. Spending big money generally will, however, buy an amplifier made from more and/or better parts. Speakers are different. Most exotic speakers are neither properly engineered nor meticulously built, as their measurements prove, yet they fre- Many of today’s most expensive speaker models use the same quently cost far more than products which are technologically off-the-shelf drivers, and other components, as products that superior and sound better. Making an enclosure from machined cost far less. Rather than providing superior performance, many unobtainium can’t compensate for a lack of engineering skill or of the highest-priced speakers available today are demonstra- the performance limitations of off-the-shelf drive elements. bly inaccurate. No loudspeaker with a couple hundred Loudspeaker drive elements are a lot better than they used to be—some technological advancements are real. While the per- dollar’s worth of component parts inside formance capability of speaker component parts has improved and the cost of these components has risen only modestly, the should cost $20,000. price tags on some models of complete loudspeakers are sim- ply absurd. The truth is that some of the world’s best loudspeakers can be purchased for far less than $20,000. For about one third that You won’t get better performance by paying these high prices. price, $6,000 to $8,000, you can buy an outstanding stereo In fact, you’ll most likely get demonstrably inferior performance speaker system that will provide demonstrably more accurate compared to better-engineered, lower-cost alternatives. The performance and better sound than the exotic systems you magazine reviewers, who present themselves as your advo- read about in Stereophile which cost as much as a luxury auto- cates laud these poorly conceived products and never mention mobile! Let me give you some examples. that better sound can be had for less. Often far less. The world’s best electrostatic speaker, the Quad 988, sells This issue of the Audio Perfectionist Journal is all about for about $6,000 a pair in the USA. The same money, about loudspeakers and we’re going to concentrate on accurate, high- $6,000, will buy a pair of Vandersteen 3A Signature speakers value speakers here. As usual, I’m going to start right off with along with a pair of Vandersteen 2WQ powered subwoofers— some bold and controversial statements. the stereo speaker system I’m listening to now.
Real Value versus Perceived Value Thiel CS6s and Dunlavy SCIV/As cost about $8,000 a pair. Pay If you believe that today’s best high-end speakers must cost $10,000 to $15,000 and you’ll get a little better performance $20,000 a pair or more, you may be pleasantly surprised to dis- from models in the Dunlavy, Thiel and Vandersteen lines. (It cover that you are mistaken. Anyone who spends that much is costs a disproportionate amount to squeeze that last five per- paying for something other than performance. cent of performance from a loudspeaker.) The top models from these manufacturers are dramatically better speakers than the Prestige and “pride of ownership” are perfectly valid reasons for prestige products costing $20,000 to $30,000 and more that get buying things, of course, but the Audio Perfectionist Journal so much press these days.
6 www.audioperfectionist.com Crisis in the Industry
A speaker system that can deliver a satisfying musical experi- This Journal is an attempt to alert consumers to this egre- ence for all but the most critical listeners, the Vandersteen 2Ce gious situation. My goal is not to sell you Dunlavy, Thiel or Signature, can be purchased for about $1,500 a pair. The 2Ce Vandersteen speakers. I want to demonstrate that truly accu- is a time- and phase-accurate, floor-standing, four-way, “baffle- rate, well-engineered and genuinely well-made speaker sys- less” design that has broader bandwidth and flatter frequency tems can be had for prices that are fully justified by the manu- response than the two models of $10,000 a pair speakers that facturing costs involved. Stereophile reviewed in the same issue, along with the 2Ce Signature. (See Watch Dog #6.) How can that be? If you are looking for the most accurate speakers available and want to spend the least amount possible to own them, you Are the manufacturers of the overpriced but poorly performing will likely choose one of these brands because these three products at the top of the speaker price heap really trying to rip manufacturers are the only major companies making time- and us off? Are the magazines aiding and abetting them? Read on phase-accurate speakers today. But you may find another and judge for yourself. brand with products that please you more. I won’t mind.
The High-End Myth You shouldn’t spend bundles of money simply because you Most “high-end” speaker manufacturers don’t actually make believe that a greater expenditure will guarantee better perfor- anything, and many “genius” designers don’t actually design mance. Examine what goes into these speakers. Then listen the “breakthrough” products they tout. Anybody with a computer and compare them to other products before you write that can buy software which will automatically design a standard big check. Ask a dealer to insert a pair of Vandersteen 3A “textbook” crossover network to integrate the output from read- Signature speakers, or a pair of Thiel CS2.3s, into his highest- ily available speaker drive elements. These crossovers can be priced audio system in place of the more expensive speakers manufactured by electronic assembly houses that specialize in he is probably demonstrating. You may be surprised to find that “stuffing boards.” Purchase some off-the-shelf drivers and put the sound actually improves as the price goes down. them in rectangular fiberboard boxes, along with these simple crossovers, and you’re a speaker manufacturer. The following articles present information about three loud- speaker brands in alphabetical order. Each company is You don’t even have to make the boxes because you can buy well established—Dunlavy for nearly ten years; Thiel and them from large woodworking firms that do nothing but manu- Vandersteen for nearly twenty-five years. Each company is facture speaker enclosures. headed by a real engineer who understands the laws of physics and knows the immutable rules of electronics. Get enough press from the influential magazines, which can be easily obtained with a big advertising budget, and you’re a Each product line is unique and each line offers a different design “genius.” Charge an exorbitant price and your product approach to solving the problems that cause most speakers to may be perceived as a technological breakthrough, especially deviate from accurate response. All the products from these if you add some advertising hyperbole and have a charismatic three manufacturers have some things in common: they sound company leader with a toothy smile and a good line of BS. A good, they’re demonstrably accurate, and they are reasonably stratospheric price tag will guarantee that your product will be priced. Products from these three manufacturers offer one thing perceived as a high quality offering even if it really isn’t. that you won’t get from most other speakers: waveform integ- rity. Think about this folks. Respectable companies like Polk and Paradigm can offer well-engineered, high quality speaker sys- The acoustical waveform that comes out of these speakers tems for a couple hundred bucks while some high-end speaker will look very much like the electrical waveform that goes in “manufacturers” ask $10,000, $12,000, even $20,000 for speak- when viewed on an oscilloscope screen. You’d expect that from ers with no more technology and only slightly higher (in the any amplifier. Why not demand the same performance from a best cases) component quality. speaker system? APJ
www.audioperfectionist.com 7 Time and Phase, Not Just a Craze
Time and Phase, Not Just a Craze by Richard Hardesty nant intervals produced by musical instruments. He described the musical scale and the importance of partials (sometimes Is loudspeaker time domain performance audible? Do time- and called overtones) in determining the characteristic sound of phase-accurate speakers provide better sound than convention- instruments. Nearly twenty-five hundred years before the oscil- al time-incoherent speakers? Harmon International’s Dr. Floyd loscope was invented, Pythagoras knew that musical tones Toole says no and I say yes, absolutely. Who can you believe? were made up of complex combinations of frequencies.
Toole and others have performed listening experiments which About one hundred fifty years ago, Hermann Helmholtz listed “prove” that average listeners cannot hear the effects of phase the three factors which distinguish musical tones. These fac- shift and time smear. I have personally demonstrated, to a far tors are force, pitch, and quality. Force describes the amplitude larger number of people, that time- and phase-accurate speak- of the vibrations which define the loudness of the tone. Pitch ers can be easily distinguished from speakers with steep filter describes the frequency of the vibrations or oscillations. Quality slopes, even when both types have comparable amplitude describes the type of vibrations or the shape of the wave- response linearity. form they create. The factor of quality allows us to distinguish between a violin, a clarinet, a flute or a human voice, even This conflict in opinions can be confusing to readers who may though all four may sound the same note (pitch) at the same not have had the opportunity to experience a properly conduct- level (force). ed presentation. Today, many assume that our ears can detect only volume The fact is that nobody can conclusively prove what someone (force) and frequency (pitch). This is clearly not the case. else can or cannot hear under all conditions. Whether others How is it possible for us to hear individual instruments in a can perceive a given phenomenon, and the relative importance large orchestra and to hear the differences in tonal character of their perceptions, is irrelevant to us anyway. What matters to between one violin and another if our ears can only detect fre- each individual is whether certain things are audible to that indi- quency and amplitude? I believe that the answer is obvious. vidual and whether those things make a significant difference to that individual. Our ear/brain mechanism can detect the shape of the waveform that moves our ear drums. Waveform shape is created by the You’ll have to decide if time- and phase-accuracy is important relative amplitude of the frequencies that make up the wave- to you. I can give you some clues about what to listen for and form and by the timing relationships between these frequen- then you can decide if what you hear is worthy of your concern. cies.
This article is filled with rhetoric about why I think that time It is clear that we can sense how far our ear drums move, domain accuracy is an important aspect of loudspeaker perfor- which allows us to detect the force (amplitude) of a tone. We mance. I believe that it is. I’ll present some logical arguments can sense the frequency of the vibrations of our ear drums to for my position and tell you what to listen for in your own exper- determine the pitch of the tone. Just how we detect the qual- iments. In this Journal we’ll look at graphs of sine waves and ity of a tone is not fully understood but the fact that we can is resultant combinations of sine waves. In the next issue we’ll clearly evident. examine graphs of the output waveforms of various loudspeak- ers and talk about sound. Helmholtz, in his groundbreaking work On the Sensations of Tone, compared the perceptions of the ear to those of the History Lesson human eye. These sensory organs have very different capabili- More than five hundred years before Christ, Pythagoras wrote ties. While the eyes can perceive an overview of simultane- the law of consonance for strings. He discovered that there was ously occurring events, the ears are limited to the perception of a physiological reason for the numerical ratios between conso- the result of these events.
8 www.audioperfectionist.com Time and Phase, Not Just a Craze
Eyes sound waves impinge on the ear simultaneously, the ear drum When a stone is dropped into a pond, concentric ripples (a moves in response to the resultant air pressure, which is the waveform) emanate from the point where the stone entered the algebraic sum of the two (or more) pressure waves. How then water. When several stones are dropped into a pond, each one do we perceive the “quality” of a tone? creates a unique set of ripples in concentric circles around the point of entry (waveforms). Periodic Waves The ear drum responds only to air pressure. Sounds are pres- The expanding circles of ripples from different stones can inter- sure waves in the air. Musical tones are complex sounds. sect, and the amplitude of the ripples can add or subtract at Musical instruments produce tones made up of many frequen- the point of intersection. Each waveform can pass through and cies. beyond the others while maintaining its individuality. The eye can perceive all the individual waveforms simultaneously and These many frequencies add up algebraically to produce a watch as they cross and continue on. sum which can be represented as a resultant periodic wave, as demonstrated by Helmholtz. Virtually everyone accepts the fact I have observed this while sitting on a cliff above the ocean that the ear perceives the amplitude and frequency of the air and when paddling my kayak in various bodies of water. Waves pressure fluctuations created by this resultant periodic wave. traveling in one direction pass through boat wakes traveling in another and both continue on as before. I believe that the perception of quality is based on the shape of the waveform. What else is there that would allow us to distin- If the crest of a boat wake meets the crest of a wave the level guish between a Steinway and a Yamaha? of the water becomes momentarily higher than either (addition). If the trough of one meets the crest of the other, the height of The shape of a resultant waveform is determined by the ampli- the larger is diminished momentarily by the level of the smaller tude and phase of the individual waves that combine to make a (partial cancellation). The eye can observe these interactions resultant wave. To help visualize this, let’s look at some simple individually and simultaneously but the ear works differently. sine waves as they appear on an oscilloscope. (Figures 1 The ear, like the kayak, is only impacted by the result. through 4 are “screen shots” taken directly from my digital oscil- loscope. The sine waves were created by signal generators and Ears the math was performed in the ‘scope. These are actual signal Sound waves in the air are like ripples in a pond. The ripples in traces, not drawings.) a pond are made up of crests where the water level is slightly higher than the average level, and troughs where the water Figure 1 shows a 100Hz sine wave and a 200Hz sine wave level is slightly lower. Sound waves are made up of crests with the higher frequency wave at half the amplitude of the where the air pressure is slightly higher than the ambient pres- lower frequency wave. This could represent the sound of an sure, and troughs where the air pressure is slightly lower.
When a crest (higher pressure) strikes your ear drum it moves inward. When a trough (region of lower pressure) arrives, your ear drum moves outward in response to the higher ambi- ent pressure inside your head. The ear drum cannot perceive sound waves propagating across the room like the eye can perceive water waves propagating across the sea. The ear can only detect the pressure fluctuations impinging on the ear drum while the brain records when they occur.
When two sound waves intersect, the pressures combine alge- Figure 1 braically just like the ripples on the water. When two or more
www.audioperfectionist.com 9 Dunlavy
Dunlavy by Richard Hardesty
Dunlavy Audio Labs was founded in 1992, making it the youngest of the three manufacturers we’ll examine in this issue. However, speaker designer and company founder John Figure 2 Dunlavy has been around quite a bit longer. He has headed instrumental tone where the harmonic is half as loud as the fun- two previous loudspeaker companies—both named Duntech. damental with the phase relationship shown (arbitrary). The first Duntech was headquartered in Texas and pro- Figure 2 shows the resultant waveform that would be recorded duced a time- and phase-correct three-way speaker system, when the 100Hz fundamental and the 200Hz harmonic are the DL-15, in the 1970s. The more recent and better-known combined algebraically, as they would be by a microphone or Duntech speaker company was an Australian firm that made the ear. If the phase relationship between the two waves is highly-regarded speakers in the 1980s and early 1990s. The altered the resultant wave will look (and sound) very different. Australian Duntechs were similar in design to the current Dunlavy products, which are produced in Colorado Springs, Figure 3 shows sine waves at 100Hz and 400Hz with the same Colorado. 2:1 ratio of amplitude. Figure 4 shows the resultant wave when these frequencies are added algebraically in the phase relation- Before entering the speaker business, John Dunlavy designed ship shown. antenna systems. He is considered an authority on certain antenna types and holds several patents on his inventions in that field.
I have known John Dunlavy casually for many years and I am familiar with his designs going right back to the first one, which I auditioned in my retail store in (I’m guessing) 1978. I was never a Dunlavy or Duntech dealer, but I have visited the Dunlavy factory in Colorado Springs and interviewed John for Widescreen Review magazine. I have watched as speakers were built and tested at the Dunlavy facility and I Figure 3 was impressed by what I saw. In this article I’ll try to describe Dunlavy’s approach to building what he feels are the most accurate speakers possible, and I’ll give you my opinions about the strengths and weaknesses of the Dunlavy designs.
Dunlavy Audio Labs makes sixteen loudspeaker models rang- ing in price from $1,500 a pair to $23,995 a pair (as of June, Figure 4 2001). We’ll concentrate on the six-model Reference Series in Waveform Integrity this article but other models, like the Cantata and Aletha, are Accurately maintaining the original shape of the waveform is similar in principle and I’ll mention how these differ. critical if you want to accurately reproduce the original sound. Waveform shape makes a difference in imaging, too. In the next Journal we’ll discuss these things and I’ll show you the The Basics signals that come out of some speakers which do maintain Dunlavy speakers are “phased array” designs. All models waveform integrity and the signals that emanate from some that have a single, centrally-located 1-inch soft-dome tweeter and don’t. APJ matched pairs of midrange drivers arranged one above and one
10 www.audioperfectionist.com Dunlavy below the tweeter. Larger models add additional drivers, which, The Cantata and Aletha models feature symmetrical arrays in most cases, are symmetrically arranged above and below the of tweeters and midrange drivers with a single, downfiring midrange drivers. Dunlavy speakers are individually measured woofer driver placed near the floor. These models perform in a and adjusted to deliver flat frequency response within narrow similar manner to comparable models in the Reference Series limits. except that they have one larger woofer instead of two smaller drivers. The downward-firing woofer allows a front baffle with All Dunlavy speakers are time- and phase-correct with first- less width. The Cantata and Aletha models have slightly less order acoustic slope transitions between driver elements and dynamic range and deliver sound that is slightly more open stepped front baffles which mechanically align the individual than what you hear from their Reference Series counterparts, in drive units. my opinion.
All models use sealed enclosure, low-Q bass loading for Phased Arrays extended, well damped bass. John Dunlavy chooses soft The “phased array” design has many advantages and some materials with high internal damping for all driver diaphragms. disadvantages, too. Advantages include controlled vertical dis- Edge diffraction is minimized by the use of a thick blanket of persion, increased sensitivity and added cone area for higher felt which surrounds the tweeter in all models and also the mid- output, especially at low frequencies. range drivers in the larger speakers. Vertical dispersion is narrowed, minimizing the effects of floor All Dunlavy speaker models share the same basic design and and ceiling reflections. While I consider “floor bounce” to be the same tweeter. Each step up from one model to the next a measurement phenomenon with little or no negative impact provides extended bass response and greater dynamic range. on sound, the multiple-woofer design helps to ameliorate this “problem.”
The $1,500 a pair SC-1/AVs are 2-way speakers with three drivers: one 1-inch tweeter and two 5.5-inch midrange units. All Dunlavy speaker models
The 2-way, $2,495 a pair SCIIs offer extended bass response share the same basic design and slightly greater dynamic range, when compared to the SC-1/AVs, by employing slightly larger 6.5-inch woofers and and the same tweeter. floor standing cabinets with greater internal volume. The woofers load the room from different locations, one near SCIIIs and SCIV/As are 3-way speakers with dual 5-inch mid- the floor and one several feet above the floor, for smoother range drivers and 1-inch tweeters. The SCIII uses two 6.5-inch in-room bass response. Doubling up on drivers increases sen- woofers arrayed one above and one below the midrange driv- sitivity by as much as 6dB because of mutual coupling between ers. SCIIIs sell for $3,995 a pair. The SCIV/A model has two driver pairs at certain frequencies. Increased sensitivity allows 10-inch woofers arrayed one above and one below the mid- driver nonlinearities to be corrected, by tailoring crossover range drivers and sells for $7,995 a pair. components, while keeping the overall sensitivity of the speaker system quite high. The SCV and SCVI are 4-way designs with symmetrical driver arrays. The $14,995 a pair SCVs have two 12-inch woofers, Higher sensitivity, combined with the greater diaphragm surface two 6.5-inch midbass drivers, two 3-inch midrange drivers and area provided by pairs of drivers covering the same frequency a single 1-inch dome tweeter in each speaker. The $23,995 range, allows the speakers to play louder for increased dynamic a pair SCVIs utilize two 15-inch woofers, two 8-inch midbass range and reduced power requirements. Double woofers move drivers, two 5-inch midrange drivers, and a single 1-inch dome more air at low frequencies allowing the use of sealed enclo- tweeter in each speaker. sures for superior bass performance.
www.audioperfectionist.com 11 Dunlavy
Disadvantages include the large enclosures required to imple- data is recorded for future reference, in case a replacement is ment the phased array designs and the fact that perfect driver required. matching is a practical impossibility. The vertical height of listeners’ ears becomes extremely critical with dual midrange drivers, too. Cone drivers in Dunlavy speakers use pulp paper-based dia- phragm materials loaded with stiffening agents like carbon fiber. Even with close inspection and matching there will be substan- The tweeters have soft fabric domes. Dunlavy chooses these tial variations in the response of two different drivers, especially diaphragm materials based primarily on impulse response test- those using paper cones. Two midrange drivers with slightly ing. different response curves will have a smoothing effect on the measured frequency response of the speaker. While a perfectly The choice of diaphragm materials is a matter of intense positioned measuring microphone “hears” only the combined debate among speaker manufacturers. Advocates of softer, output of both drivers in a pair, the ear may detect subtle differ- highly damped materials, like those used by Dunlavy, point ences. Even with perfectly matched drivers, the listeners’ ears out that diaphragms with increased rigidity often have large must be exactly equidistant from each of the midrange units, or resonant peaks and exhibit ringing. Advocates of stiffer, more time smear and some comb filtering will occur. rigid diaphragm materials claim that softer diaphragms produce higher distortion because these diaphragms fail to perform as The enclosures required for the sealed box, phased array perfect pistons and “break up” at relatively low frequencies. designs are larger than what would be necessary for conven- Here are both sides of the story. tional multiway speakers using single drivers for each frequen- cy range. Larger enclosures contribute more unwanted energy to the sound due to larger box panels and larger baffles around Diaphragm Materials drivers. Softer diaphragm materials, like fabric and reinforced paper pulp, have high internal damping and may exhibit few reso- Larger panels store more energy and resonate at lower fre- nant peaks within the usable bandwidth and to frequencies quencies. Larger baffles reflect mid and high frequencies. Both well above the normal operating range of the driver. Resonant factors cause time smear and tend to make speakers sound peaks which do exist will usually be low-Q and well damped. slightly more “closed in” and “boxy.” Impulse response tests on drivers with softer diaphragms will not exhibit the overshoot and ringing that can often be seen Dunlavy uses felt around the tweeters to minimize baffle reflec- when testing drivers using stiffer diaphragm materials. tions and diffraction, and the speaker cabinets are heavily braced with thick panels to control resonance. These attempts However, laser interferometry shows that softer diaphragms to minimize the problems of large enclosures are not entirely tend to break up at lower frequencies. Some soft-dome tweet- successful, in my opinion. The smaller Dunlavy models sound ers, for instance, start to break up (perform like imperfect pis- more open, spacious and detailed than the larger ones, to me. tons) at frequencies below 10kHz.
When a radiating diaphragm begins to perform less like a per- Drivers fect piston due to “break up,” distortion rises and resolution The drivers used in the Dunlavy speaker systems are sourced is diminished. Soft materials absorb some energy that might from Vifa/Scanspeak. These are among the highest quality driv- otherwise become sound, compressing dynamic contrasts and ers available today, despite what you might hear elsewhere. obscuring subtle details in the signal.
All drive elements feature cast alloy frames and ceramic mag- Stiffer diaphragms, made from materials like polyamides nets. Every driver used in a Dunlavy speaker is individually (Kevlar), metal alloys, and ceramic (actually anodized alumi- tested and each one must meet strict performance standards. num), may act more like perfect pistons throughout the normal Drivers are closely matched and the individual performance operating range of frequencies and well above. But at some fre-
12 www.audioperfectionist.com Dunlavy quency, even a stiff diaphragm will break up and fail to perform Crossover Networks as a perfect piston. When a stiff diaphragm reaches its first The crossover networks in all Dunlavy speakers are engineered resonance or break-up frequency, a sharp resonant peak with a to provide frequency and phase compensation for the drive ele- high-Q is often observed. ments and to provide first-order acoustic transitions between drivers. All crossover components, internal wiring and circuit Within their useful bandwidth, stiff diaphragm materials may boards are of high quality. exhibit lower distortion and less coloration allowing higher reso- lution of minute details in the signal. When stiffer diaphragm The high sensitivity provided by pairing midrange and woofer materials do “go off” there is a tendency for the high-Q reso- drivers allows the crossover networks to provide lots of fre- nance to ring over a period of time due to a lack of internal quency response and phase response tailoring, while main- damping. taining high overall speaker sensitivity. Response tailoring is accomplished by adjusting component values on the crossover Which diaphragm material is better, a highly damped substance boards during design and testing. Individual crossovers are par- which breaks up in the pass band or a stiffer material that pro- tially assembled and tested, and then matched with a speaker duces a ringing resonance out of the pass band? for final tweaking.
Actually both faults, cone break-up and ringing, cause time Each speaker is placed in an anechoic chamber and measured smear due to the storage and delayed release of energy. using this basic crossover. Crossover component values are Strong arguments can be made for each of these choices. individually adjusted by a technician until the desired perfor- Dunlavy uses soft materials exclusively, Thiel uses aluminum mance objectives for the complete speaker are met. (almost) exclusively, and Vandersteen uses several different materials for different frequency ranges. You’ll have to go and listen to discover which you prefer. Quality Control All speaker components are quality selected and inspected at every phase of construction. Individual drive units are tested, Enclosures selected, matched in pairs for each speaker and matched in Dunlavy enclosures are rectangular boxes (with the exception fours for each speaker pair. Individual crossover components of the Alethas, which are hexagonal as viewed from the top). are selected and crossover assemblies are tested before final All models have steps in the front baffles to temporally align the adjustments are performed during anechoic testing. Enclosures drivers. (I can’t say “time align” because E. M. Long has trade- are inspected for mechanical integrity and fit & finish. marked those particular words and effectively taken them out of the English language.) Every completed Dunlavy In all Dunlavy speakers, the tweeter is mounted on a baffle, which is stepped back from the mounting baffles on which speaker is placed in an the midrange drivers are mounted. Midrange driver baffles are stepped back from the surface on which the woofers are anechoic chamber mounted, on models which utilize separate woofer drivers. and measured. Enclosure walls are made from MDF and are internally braced. The larger models have 1-inch thick panels and are very Every completed Dunlavy speaker is placed in an anechoic heavy. Dunlavy cabinets are nicely finished with furniture grade chamber and measured. Crossover components are individually veneers and are available in a variety of woods. They look adjusted to produce the desired response characteristics. Each quite traditional—wood veneered, rectangular boxes with grille speaker and crossover assembly must equal or exceed the fre- cloths covering the drivers—and most models are quite large. quency response and phase response specifications that
www.audioperfectionist.com 13 Dunlavy speakers do not image as well as Thiels and Vandersteens,Dunlavy
are guaranteed for that particular model. Speakers are built and The smaller Dunlavy models have compromised bass output tested in pairs, and each pair is matched within narrow limits. compared to similarly priced products using vented enclosures. Vents allow smaller speakers to deliver higher bass output lev- els at some frequencies. Performance graphs are generated for all Dunlavy speakers. These graphs demonstrate that the speakers you receive meet Dunlavy uses soft dome tweeters and comparatively large or exceed the advertised performance specifications. midrange drivers (in most models), which deliver consistently smooth and musical sound, but some critical listeners may long for higher definition and more midrange detail along with a Strong Points more extended and airy top end. Model for model, the Dunlavy speakers generally have higher sensitivity and greater dynamic range than time- and phase- accurate speakers using fewer drivers. They will play somewhat Measurements louder and require somewhat less amplifier power to perform at Impulse response is the Holy Grail to John Dunlavy so it’s not maximum capability. surprising that his speaker designs perform well when tested with the MLSSA system. And perform well they do. All Dunlavy models feature low-Q sealed enclosures. Sealed enclosures provide tighter bass control along with superior bass Dunlavy speakers deliver impeccable measured performance definition, deeper extension to well below the nominal cut-off on all tests except cumulative spectral decay. Frequency frequency, and greater phase linearity with less group delay response of the smaller models is guaranteed to be flat within at low frequencies, when compared to most vented designs. ±1.5dB. The larger models are guaranteed to be flat within Dunlavys have flatter phase response at low frequencies than ±1dB. Impulse and step response graphs look like textbook time- and phase-accurate speakers which use vented bass examples of ideal time domain performance. Compare these systems. measurements to those of the speakers discussed in Journal #5.
Negative Considerations Remember, perfect measurements do not guarantee perfect The large rectangular enclosures of the Dunlavy speakers, par- performance. In fact, sonic compromises may be required to ticularly the bigger ones, contribute some sound of their own. produce flawless graphs. In Audio Perfectionist Journal #8 Big boxes sound like big boxes, to me. we’ll discuss the things that measurements don’t show and I’ll tell you how, in certain circumstances, measurements can actu- Dunlavy speakers do not ally be misleading. APJ image as well as Thiels and Vandersteens.
The effects of wide baffles can be heard when comparing the smaller models to the larger ones within the Dunlavy line. The sonic price you pay for the extended bass response and increased dynamic range of the larger models is a slightly less open and spacious sound.
Dunlavy speakers do not image as well as Thiels and Vandersteens, in my opinion. Paired midrange drivers tend to blur image focus and response differences between these driver pairs may produce some vertical image wander.
14 www.audioperfectionist.com Dunlavy
Dunlavy SC III
www.audioperfectionist.com 15 Dunlavy
Dunlavy SC III
16 www.audioperfectionist.com Dunlavy
Dunlavy SC IVA
www.audioperfectionist.com 17 Dunlavy
Dunlavy SC IVA
18 www.audioperfectionist.com Thiel
Thiel the high-end segment of the market. Thirty-five employees work by Richard Hardesty in a modern 40,000 square foot factory, making twelve loud- speaker models ranging in price from $1,450 a pair to $13,500 Thiel Audio Products began shipping the Model 01, Jim Thiel’s a pair (as of June, 2001). first commercial speaker design, in 1977. Shortly thereafter, in 1978 or so, I became a Thiel dealer and began selling their We’ll concentrate on the six-model floor-standing CS Series in bookshelf Model 02 and the first time- and phase-accurate this article but other models, which are aimed more at home Thiel speakers, the 03 and 04. I have closely followed the theater use, will be discussed in future Journals. development of the Thiel Audio loudspeaker line since the beginning, and maintained an ongoing friendship with designer Jim Thiel and company president Kathy Gornik for more than The Basics twenty years. Thiel speakers are “Coherent Source” designs with sloping front baffles, which temporally align the individual drive elements. I sold, installed and repaired Thiel speakers in the 1970s and Thiel uses unique dynamic drivers with “underhung” voice coils 1980s and I have had much experience with later models in the and metal diaphragms. Most Thiel drivers are designed and ‘90s through the present. (What do we call this decade? The manufactured in the Thiel factory including all the drivers in the 00s?) I reviewed the CS7 for Widescreen Review and used models CS2.3, CS6 and CS7. that speaker system to evaluate more than fifty amplifiers for my series on multichannel amplifiers, which ran in WSR over a period of several years. Thiel speakers are
Widescreen Review has a complete Thiel home theater speak- “Coherent Source” designs er system based on four Model CS6s. I’ve spent many hours listening to this system and I’ll be reviewing it later this year. I with sloping front baffles. personally installed the latest drivers in the CS6s at WSR so I’ve had hands-on experience inside some of the newest mod- Thiel speakers deliver flat frequency response within narrow els in the Thiel line. limits. All Thiel speakers are time- and phase-correct with first- order acoustic slope transitions between drive elements. The In preparation for this article, my wife Paula and I visited the sloped and contoured front baffles, which mechanically align Thiel facility in Lexington, Kentucky, and spent a couple of days the individual drive elements relative to the listener, also mini- with Jim and Kathy. I wanted to see their new factory and to mize edge diffraction and baffle reflections. interview Jim so that I could accurately convey his design phi- losophies in print. That interview will be published in WSR later All CS Series models have vented bass loading utilizing either this year and I’ll try to communicate what I learned about the a port or a passive radiator. Jim Thiel chooses aluminum alloy Thiel designs in this article. for nearly all driver diaphragms.
Jim Thiel is one of the most innovative and technically astute Each Thiel model is uniquely designed to provide the high- engineers that I have ever met. I’ve learned a lot from him over est definition and the broadest bandwidth possible at its price the years and the recent visit to Kentucky was highly informa- point. Each step up in price buys an increase in dynamic range, tive. When I ask Thiel a question I get a thoughtful answer, not slightly higher resolution and slightly smoother response. Low marketing rhetoric. When I ask about a competitor’s designs, bass capability is improved with each model step. Jim calmly expresses his opinions and tells me why he has chosen to do things differently. The Thiel CS Series The Lexington factory is an impressive facility. Thiel Audio The small, floor-standing CS.5 is a 2-way vented (ported) Products is one of the largest manufacturers of loudspeakers in design with a 6.5-inch paper-composite woofer and a 1-inch
www.audioperfectionist.com 19 Thiel metal dome tweeter. The CS.5s sell for $1,450 a pair (slightly When even the best off-the-shelf drivers failed to meet all of more on the west coast). The step up model CS1.5 has a simi- his demanding requirements, Jim Thiel decided to design and lar driver complement but uses a much more costly woofer fea- manufacture his own drive elements. All the drivers used in the turing an aluminum cone. The CS1.5s employ a passive radia- CS2.3, CS6 and CS7.2 models are custom units, engineered tor in place of the ducted port seen on the CS.5 model. CS1.5s for their specific purpose and manufactured by Thiel. sell for about $2,200 a pair. Thiel Audio staff members have designed and fabricated cus- The models CS2.3, CS3.6 and CS6 are 3-way speakers. The tom tooling and specialized automated machinery to make $3,600 a pair CS2.3 has a unique 3.5-inch aluminum cone mid- the drive units. This speeds development time and allows the range and a coaxially mounted 1-inch aluminum dome tweeter, designer to test new ideas almost immediately. Jim Thiel can along with an 8-inch aluminum cone woofer and a passive dream up a new cone profile and have a prototype driver, utiliz- radiator. ing the new diaphragm shape, ready for testing that same day.
The CS3.6, which costs about $4,300 a pair, uses a 1-inch A special tool can be produced by the staff CAD/CAM specialist aluminum dome tweeter, a 4.5-inch double cone midrange, and a cone can be fluid-pressure formed using this tool in the a 10-inch aluminum cone woofer and a passive radiator. The Thiel Audio prototype shop. The new cone can then be assem- $7,900 a pair CS6 has a 1-inch aluminum dome tweeter coaxi- bled into a complete driver in the factory and hand carried to ally mounted in a 4-inch aluminum/polystyrene/aluminum sand- the laboratory for testing. wich cone midrange, with a 10-inch aluminum cone woofer and a passive radiator. Some Thiel driver units are remarkably unique. The coaxial midrange/tweeter unit in the CS2.3 features two concentric The flagship CS7.2 is a 4-way design utilizing a coaxial mid- metal diaphragms driven by a single voice coil with a mechani- range/tweeter unit with a 3-inch aluminum sandwich cone cal crossover which decouples the larger diaphragm at higher midrange diaphragm and a 1-inch aluminum dome tweeter. frequencies. A 6.5-inch lower midrange/upper-bass driver with an alumi- num cone and double magnet motor system covers the range between the midrange element and the 12-inch aluminum cone The coaxial midrange/tweeter woofer, which is complemented by a passive radiator. CS7.2s cost $13,500 a pair. drivers in the CS6 and CS7.2 have sandwich cones. Drivers The drivers used in most Thiel speaker systems are designed The coaxial midrange/tweeter drivers in the CS6 and CS7.2 and manufactured in-house by Thiel. Some drivers used in the have sandwich midrange cones made from two outer skins of smaller models are currently sourced from Vifa/Scanspeak. All aluminum with a layer of polystyrene in between to provide the drive elements feature cast alloy frames. The woofers and mid- shallow depth required for an ideal tweeter environment (the range drive units utilize huge ceramic magnets to concentrate midrange cone surrounds the tweeter in a coaxial driver) along flux in the extended magnetic gaps, and some of the coaxially- with the steeper cone angle required for rigidity. mounted tweeters use more compact and powerful rare earth magnets. The tweeters in these drivers have unique neodymium magnet structures. Small, powerful magnetic systems are required to First-order speakers require exceptional performance from drive provide high flux density in the extended gaps, and coaxially units which must exhibit outstanding frequency response and mounted tweeters must have compact overall dimensions. phase performance over a much wider bandwidth than the driv- ers used in conventional speakers with steep-slope filters. Jim Thiel prefers to use stiffer, aluminum material for virtually all driver diaphragms. He favors the use of motor systems utiliz-
20 www.audioperfectionist.com Thiel ing a short voice coil in a long magnetic gap for reduced distor- used in the CS6 and CS7.2 models have a layer of polystyrene tion. He likes coaxially mounted drivers for the upper frequency bonded to the aluminum diaphragms, which helps to damp the ranges to minimize lobes in the dispersion pattern and to make metal layers. vertical listening height less critical. All of these preferences are controversial and I’ll try to present the basic arguments for both sides of each issue. Those who favor softer diaphragm materials with more internal damping point to the ringing which can be observed on impulse response tests of metal diaphragm drivers, and speaker sys- Metal Diaphragms tems based on these drivers. They contend that the sharp reso- The controversy regarding softer versus stiffer diaphragm nant peaks exhibited by metal diaphragms produce undesirable materials was discussed in the article about Dunlavy speakers audible effects even when they are “notched out” by electrical in this issue. Stiffer diaphragm materials, like aluminum, gener- filters. They point out that resonant behavior can be excited by ally perform as near-perfect pistons over a wide range of fre- harmonically related frequencies well below the frequency of quencies and then exhibit a sharp resonant peak in response. resonance. Ringing may occur due to a lack of internal damping.
Underhung Voice Coils Jim Thiel feels that his aluminum diaphragm drivers deliver The voice coil is the moving element in the motor system of a lower distortion, less dynamic compression and higher resolu- loudspeaker driver. The voice coil becomes an electromagnet tion than comparable drivers with diaphragms made from paper when energized by the signal current from the amplifier. or plastic materials, over the range of frequencies which they are called upon to produce in his speaker designs. The alternating electromagnetic field in the voice coil reacts to the fixed magnetic field created by the permanent magnet, front He can design the diaphragms so that the first resonant peak and back plates and the pole piece which make up the non- occurs at frequencies well above the range of use and Thiel moving parts of the driver motor system. This magnetic reaction feels that these out-of-band resonances can be compensated causes diaphragm movement because the voice coil is coupled for in the crossover design. The sandwich midrange cones directly to the diaphragm.
Thiel short coil, long gap woofer.
www.audioperfectionist.com 21 Thiel
The diaphragm is free to move because it is attached to the Coaxial Drivers driver frame by two springs—the spider near the voice coil and The best way to simulate a point source radiator is to make the the surround at the perimeter of the diaphragm. individual radiating elements truly coincident. Coaxial midrange/tweeter drivers are particularly advantageous in first- The voice coil has inductance and inductive reactance var- order speakers where the large overlap between driver operat- ies with voice coil position because the amount of iron in and ing ranges shrinks the vertical dispersion window and makes around the voice coil varies as the coil moves back and forth— the relationship between speaker tilt back and listener ear toward and away from the permanent magnetic structure. The height very critical. There are, however, many problems with asymmetrical variation in coil inductance can distort the output coaxial drivers. waveform. Mounting a dome tweeter in the center of a midrange driver Placing a short voice coil in a long magnetic gap eliminates has deleterious effects on the performance of the midrange much of this distortion by making inductive reactance more unit, and a moving midrange cone is a poor environment for a linear. Copper sleeves over the pole piece and copper shorting tweeter. A conventional midrange cone can horn-load a coax- rings around the base of the pole piece help to linearize the ially-mounted tweeter and a deep midrange cone will create a magnetic field and eliminate eddy currents which develop in the iron structure of the motor system. These copper parts bring about an additional reduction in distortion. Thiel uses all these distortion-reducing techniques in his custom drive elements.
Huge magnets are required to create an intense magnetic field in the extended gap, and the added cost of the enormous magnet structure is the only drawback to the short coil/long gap designs, according to Thiel.
Thiel uses all these distortion-reducing techniques in his custom drive elements. Thiel CS 6 coaxial midrange/tweeter drive unit. Those with opposing views argue that the short voice coil length concentrates heat in a smaller area which affects reli- ability and dynamic range, and that shaped pole pieces can cavity around the tweeter and add a cavity resonance which effectively reduce distortion to levels that are nearly as low as may affect response. A moving midrange cone surrounding a those achieved with short coil/long gap designs. tweeter may modulate the output of that tweeter and vice versa.
Conventional coaxial drivers don’t perform very well in objective Jim Thiel counters these arguments by pointing out that the tests or subjective listening experiments. (To hear examples increased mass of iron surrounding the coil effectively deals of this, go listen to KEF speaker models using their UniQ coax with heat build-up and that his measurements demonstrate drivers, or to Tannoy models using their Dual-Concentric driv- distortion figures that are much lower than the levels attainable ers.) Thiel has attempted to overcome these problems with with shaped pole pieces alone. unique coaxial driver designs. (To which you should also listen for comparison.)
22 www.audioperfectionist.com Thiel
Thiel’s midrange cones are extremely shallow on the surface Thiel speaker enclosures are made from heavily braced 1-inch surrounding the tweeter. In fact, the forward-facing contour of thick MDF material. All Thiel cabinets are unusually solid and these cones approximates the contour of the mounting flange acoustically inert. The contoured front baffles on the smaller surrounding the popular Vifa dome tweeter used in some Thiel models are made from machined MDF material that is up to models. 3-inches in thickness. Baffles on the CS6 and CS7.2 models are made from a cast mineral/polymer material which has a much higher compression strength than MDF. This shallow cone shape is not very rigid and that’s where the sandwich construction comes in. Thiel’s midrange cones are actually constructed from two aluminum skins separated by a Thicker baffles make better sound but increasing the MDF layer of polystyrene. This three-layer laminate has a shallow baffles beyond 3-inches in thickness failed to produce benefi- contour facing forward and a steeper contour facing rearward to cial results. The limiting factor proved to be the compression increase rigidity. The styrene helps to damp the aluminum, too. strength of the material itself.
The Thiel cabinets are much To overcome this limitation, Thiel developed a unique mineral/ polymer material that is similar to the cast granite materi- more than structurally inert; they als used for kitchen counter tops. This material is cast and machined in-house but the high cost prohibits its use on all but are among the most beautiful in the top models in the Thiel line (CS6, CS7.2). the industry. The Thiel cabinets are much more than structurally inert; they are among the most beautiful in the industry. The graceful, The advantage of this construction technique is that the front sloping design and the flawless wood finish make each speaker of the midrange cone becomes a nearly ideal shape to sur- an elegant esthetic statement. Thiel speakers are constructed round the tweeter. However, the midrange cone is still moving like fine furniture by master craftsmen. and that movement may modulate the output from the tweeter somewhat. Thiel buys entire flitches of hardwood veneers. A batch of veneer obtained from a single tree will be selected and Enclosures matched for grain and color so that each speaker in a pair You could duplicate the enclosures of many “high-end” speak- (or more) will be esthetically complementary. MDF enclosure ers using only a table saw and a router, but you couldn’t build panels are veneered inside and outside prior to machining. All Thiel speaker enclosures in your garage, regardless of your the parts for a pair of speaker enclosures proceed through the woodworking skills. factory together and are assembled into finished pieces of fine furniture as a matching set. The craftsmen who build these cabinets are rightfully proud of their work. Even a master craftsman couldn’t create the close-tolerance, compound angles and elaborate shapes required without the use of specialized multiaxis, computer-controlled woodworking Thiel will build speakers with virtually any wood veneer and machinery. Thiel has several such machines which are pro- with just about any finish and they will match cabinets to a cus- grammed in-house. These same computer-controlled machines tomer’s specifications. Unusual materials and finishes can cost are used to create the tooling that aids the manufacturing and a lot more but the result is a truly custom product. assembly processes for the enclosures and the tooling that is used to form the driver diaphragms.
www.audioperfectionist.com 23 Thiel
Crossover Networks Strong Points The crossover networks in all Thiel speakers are engineered Thiel builds some of the most elegant and thoroughly engi- to provide frequency, impedance and phase compensation for neered loudspeakers available at any price. Thiel makes every- the drive elements and to provide first-order acoustic transitions thing (in the top models) and is not hindered by the perfor- between drivers. The crossovers also compensate for the out- mance limitations of off-the-shelf drivers. Thiel speakers have of-band resonant peaks inherent in metal diaphragm drivers. extremely low distortion and high resolving power. They are demonstrably accurate in both frequency and time domains. Thiel crossovers are unusual in that they are hard-wired rather than built on circuit boards. Component parts are of exceptional quality. Custom polypropylene and polystyrene capacitors are Negative Considerations utilized along with air-core inductors wound with low-oxygen Thiel speakers cannot be biwired or biamped because they copper wire. Solid-conductor copper wire with Teflon insulation have only a single set of inputs. Beautiful cabinets cost money is used throughout the crossovers and between the crossovers and Thiels are relatively expensive. (If you visit the factory, as and drivers. Silver solder is used exclusively. I have, you’ll see that the prices are more than reasonable for the quality offered.) Vented bass systems cannot equal the per- formance of sealed boxes, in my opinion, but subwoofers can Quality Control change all that. Cabinets are inspected at every stage of construction and that attention to detail will be evident to anyone who visits a Thiel The CS series Thiels have low nominal impedance (dipping to dealer and views these elegant speakers. Thiel speakers can well below 4Ω in some cases) and are only moderately sensitive. be esthetically compared to the finest furniture pieces. Speaker They will require powerful, stable amplifiers for best perfor- pairs match perfectly. mance. They won’t play as loud as speakers with higher sensi- tivity and/or crossovers with sharper transitions between drive elements (steeper-slope crossovers). Crossover components are individually inspected and selected. Completed crossover networks are tested and matched. Each The most controversial aspects of the Thiel designs are, in crossover must pass a computerized QC test before it is my opinion, the metal diaphragms and the coaxial midrange/ assembled into a speaker system. tweeter drivers. Thiels and Dunlavys certainly sound different. You’ll have to listen to determine which is more pleasing to you.
Drivers are inspected at each stage of construction and then individually tested and matched. Thiel magnetizes the woofer Measurements magnets after the drivers are assembled because these huge The Thiel response graphs demonstrate outstanding perfor- magnets would be otherwise unmanageable. Individual drivers mance in all areas. Frequency response is quite linear. (Note are performance tested before being incorporated into speaker that the Thiel graphs have 5dB increments between lines.) The systems. Step and Impulse Response tests show time- and phase-coher- ence. Some minor ringing is apparent and this is probably due to the metal diaphragm materials used. Each completed Thiel loudspeaker is sweep tested to reveal any buzzes or other mechanical flaws. Then each speaker is Ringing is an energy storage phenomenon that causes time placed in an automated anechoic chamber and computer tested smear. The debate is about whether this minor ringing is a rea- for frequency response and phase response. Each speaker sonable trade-off for the increased resolution provided by stiffer must match the performance of a reference standard for that diaphragm materials. There will be more about this in the next model, within narrow limits. issue. APJ
24 www.audioperfectionist.com Thiel
Thiel CS2.3
www.audioperfectionist.com 25 Thiel
Thiel CS2.3
2 6 www.audioperfectionist.com Thiel
Thiel CS6
www.audioperfectionist.com 27 Thiel
Thiel CS6
28 www.audioperfectionist.com Vandersteen
Vandersteen talented designers, Jon Dahlquist and Richard Vandersteen, by Richard Hardesty suspected that the open sound of the Quads might be attribut- able to the fact that they had no reflective baffle surrounding In the late 1950s, the original Quad ESL electrostatic loud- the drive elements. They were partially correct. speaker was available for purchase only in the UK. It took near- ly a decade for Quads to appear at retail locations in the United Dahlquist was the first major speaker manufacturer to produce States. Stores selling Quads were few and far between and the dynamic loudspeakers with small baffles (one eighth wave- speakers were available only in limited quantities. length, I believe) around all drivers except the woofer. The original DQ-10 was introduced in the mid-1970s and it was a Serious audiophiles managed to find them. I bought a pair revolutionary design. from Jonas Miller Sound in Beverly Hills, California, in the early 1970s. Many speaker designers of the day were listening to With the grilles in place, the Dahlquist DQ-10 looked almost Quads, too. They were trying to figure out how to make a prac- exactly like a Quad ESL. Under the Quad-styled grilles, the tical loudspeaker that could sound as open and natural as the DQ-10 had a 10-inch Advent woofer in a rather conventional Quad ESL. sealed enclosure that made up the lower two thirds of the speaker. The other three drivers in this four-way system were The Quad ESL (sometimes called the Quad ’57) was a 2-way, each surrounded by small baffle boards, made from Masonite, three element electrostatic design with temporally aligned pan- which were attached by brackets to the top of the woofer enclo- els. (The electrostatic tweeter panel was centered between sure. two electrostatic woofer panels and set back from the plane of the woofers.) It was the first time- and phase-correct multiway Upper range drivers included a 5-inch Philips midrange, a speaker, so far as I know. The Quad could pass a pretty good 1-inch dome tweeter and a Motorola piezo super tweeter. semblance of a square wave and offered midrange fidelity that These drivers were staggered so that the voice coil termina- was simply unrivaled in this era. tions were lined up, in an early attempt at time—oops, I mean temporal—alignment. The Quads were far from perfect, however. They focused an image in a spot just big enough for a single head. They could play just loud enough to produce chamber music at realistic ...beneath his modest facade is sound pressure levels for a listener seated no more than six or seven feet away. one of the sharpest minds that
Quads had limited bandwidth with very little output capabil- I’ve ever encountered... ity in the low bass or high treble regions. They were fragile—I sold mine when my stock of tweeter panels was depleted— The drivers were not arranged in a vertical array. Due to its and they presented a difficult load for the amplifiers of the steep-slope crossover filters and primitive attempts at driver day. The reward, for the few enthusiasts who could tolerate alignment—there were no FFT analyzers then—the DQ-10 was these impracticalities, was glorious sound and an immensely not a time- and phase-accurate speaker system. But it was a satisfying musical experience. After parting with my Quads, I major advance in dynamic speaker design. struggled for many years before I was able to fully duplicate this level of musical satisfaction. Vandersteen went a step farther to eliminate what he refers to as “baffle distortion.” The original Vandersteen Model 2 speak- ers had virtually no baffle at all. The first Model 2 had an 8-inch High Standards woofer mounted in a trapezoidal (as viewed from the top) enclosure with no baffle extending beyond the sides of the bass The Quad ESL speakers set a high standard for the audiophile driver. A 10-inch rearward facing passive radiator shared this speaker designers of the day. Many labored to create practi- same enclosure. cal loudspeakers with the open, natural sound of Quads. Two
www.audioperfectionist.com 29 Vandersteen
The dome midrange and dome tweeter drivers were mounted While a big percentage of the price of competing speaker in free air, supported by a backbone structure attached to systems is invested in the enclosures, most of the price of the woofer enclosure. This was the first “baffleless” Aligned Vandersteen speakers is devoted to the components inside. Dynamic loudspeaker and Vandersteen has been building These components, even in the lowest-priced models, are speakers based on the same basic principle ever since. among the finest available regardless of price.
Although current Vandersteen baffleless speakers are also Vandersteen likes to refer to the $3,500 a pair Model 3 time- and phase-correct, the original Model 2 was not com- Signature as “a $6,000 speaker system in a plain wrapper.” I’d pletely phase coherent. go farther and say that the Model 3 Signatures can be favor- ably compared to any speaker system selling for $8,000 a pair Minimizing baffle reflections proved to be only a partial solution. or more. There are negative aspects to this “bargain” pricing While both Dahlquist and Vandersteen had successfully man- (but only for the manufacturer not the consumer). aged to reduce the negative audible effects produced by reflec- tive baffles surrounding the drive elements, their early speakers Because Vandersteen speakers are so inexpensive, they have were not time- and phase-correct. failed to get much attention from the influential magazines or from audiophiles who believe that high cost is synonymous The Dahlquist DQ-10s and the original Vandersteen Model 2 with high quality. I’m going to provide slightly more detail when speakers sounded more open and natural than conventional describing Vandersteen products in an attempt to dispel this box loudspeakers but neither could match the transient speed false notion. and transparency of the time- and phase-accurate Quad ESL. These relatively inexpensive speakers are built to qual- ity standards that equal or exceed any in the industry, and Because his speakers cost less Vandersteen speakers contain many innovative design features which can’t be obtained anywhere else at any price. than other high-end brands, Shane Buettner and I visited the Vandersteen factory in early Vandersteen sells lots more July in preparation for this article and a review that I had planned for Widescreen Review magazine. (It appears at the of them. time of this writing that Shane, rather than I, will be doing that review.) Vandersteen made a giant leap toward closing the gap with the introduction of the Model 2A—the first time- and phase-correct Vandersteen Audio employs twenty-two people in the small Vandersteen speaker—at the end of 1977. central California town of Hanford. The factory is located on the outskirts of the central business district in a 50,000 square foot, Today, Vandersteen Audio makes fifteen time- and phase- free-standing building built and owned by Richard Vandersteen. accurate loudspeaker and subwoofer models including many designed primarily for home theater use. In this article we’ll The facility is as low-key as the owner. There is no sign on the concentrate on the four floor-standing, full-range models num- building other than the address. There are no plush listening bered 1, 2, 3 and 5. (There used to a Model 4 but, alas, it is no or meeting rooms where visitors can be entertained and there more. There will be a Reference Monitor, priced between the 3 are no fancy offices for the sales and management staff. In fact and 5, but it’s not available yet.) there is no sales and management staff. This is a family busi- ness run by the patriarch and manned by a hand-picked crew, Because his speakers cost less than other high-end brands, which includes two sons and a daughter. Vandersteen sells lots more of them. The lower price tags are achieved by minimizing the costs of expensive, furniture-grade Richard Vandersteen likes to describe himself as “just a truck enclosures on all but the flagship Model 5. driver from Hanford, California,” but beneath his modest facade is one of the sharpest minds that I’ve ever encountered. He and
30 www.audioperfectionist.com Vandersteen
I have been good friends for nearly twenty-five years and I was The faces of these subenclosures are staggered to temporally one of the very first Vandersteen dealers, so I am admittedly align the drive units. Adjacent structures are rounded and cov- biased. The man’s engineering talent is quite extraordinary and ered with felt to eliminate any reflective surfaces or diffractive his ingenuity has led to some breakthrough technology. edges around the drivers. Woofer enclosures are trapezoidal He studied advanced Fast Fourier Transform measurement in shape (as viewed from above) so that no reflective surfaces techniques with the late Richard Heyser at Cal Tech, but extend to the sides of the low frequency drivers. Grille cloths Vandersteen had already pioneered the use of FFT analysis in are wrapped around dowels placed well away from the enclo- speaker design some years before. Vandersteen Audio was the sure assembly for complete sonic transparency. first speaker company to employ FFT analysis, using an early Gen Rad 2512 instrument for research and development. The driver outputs are blended with first-order acoustic slope, impedance compensated crossover networks with battery- Today, Vandersteen also employs the TEF and MLSSA instru- biased capacitors in the flagship models. ments along with his keen ears. I’ve shared listening sessions with many key figures in the audio industry (including the three designers profiled in this issue of the Journal) and Richard Vandersteen uses a variety of diaphragm materials depending Vandersteen is among the most aurally perceptive. (I’m preju- on the application. Various metal alloys are utilized for subbass diced here too, because he and I share similar tastes in music drivers and tweeters. Upper bass, and midrange frequencies and equipment.) are reproduced by mica-filled polymer cones with proprietary profiles. The upper bass/lower midrange driver in the Model 5 Vandersteen has developed many unique technologies which has a Kevlar-reinforced cone with a mica-filled polymer skin on are offered in various speaker models. The patented reflection- one side. free midrange driver used in the Model 3 and Model 5 speaker systems, the true push-pull subwoofer driver used in the Model 5, the baffleless enclosure designs used throughout the line, The patented reflection-free and the multilayer constrained-layer construction of the Model 5 head module are examples. midrange driver and the true
Vandersteen Audio, in the early 1980s, was the first company push-pull subwoofer driver are to build subwoofers that operate primarily below system reso- nance. These subwoofers are essentially aperiodic in the pass unique Vandersteen designs. band. Bass loading varies from model to model. The Model 1 has a Vandersteen Class B subwoofer amplifiers have a number of bass transmission line. The Models 2 and 3 have dual active circuit innovations. Battery-biased crossover networks are still drive units (8-inch and 10-inch) that share a common sealed unique in the industry. Forerunners of the current Vandersteen enclosure but operate over different frequency ranges. This “open frame” midrange and push-pull subwoofer drivers date unique bass loading system performs like a theoretically perfect back to the early 1980s. Aligned Dynamic baffleless designs QB3 vented system over much of its range. date back to the late 1970s. The Model 5 has a built-in powered subwoofer with a remark- able driver that features a true, symmetrical push-pull motor The Basics system. The bass enclosure of the Model 5 is sealed. Midrange Vandersteen speakers are Aligned Dynamic “baffleless” and tweeter drivers are transmission line loaded in all designs. That means that there are virtually no baffles extend- Vandersteen models. ing to the sides of the drive elements. Each driver is mounted in an individual subenclosure with a front face that is just large Each Vandersteen model is a full-range speaker system. enough to contain the mounting flange of that driver. Stepping up in the model range buys higher resolution, primar-
www.audioperfectionist.com 31 Vandersteen ily, along with smoother response and some bass extension. about 100Hz. This subwoofer interface is similar to the method Adding Vandersteen subwoofers improves the midrange defini- used with the Vandersteen 2WQ subwoofer (see Audio tion of all models except the Model 5, which has an integral Perfectionist Journal #2). The Model 5 subwoofer system uti- powered subwoofer. lizes a 400-watt internal amplifier (in each speaker) and offers multiband parametric equalization and adjustable “Q.”
Vandersteen Full-Range Speakers Midbass in the Model 5 is handled by a transmission-line- The Vandersteen Model 1C is a 2-way, floor-standing, truly loaded 7-inch driver featuring a kevlar-reinforced cone with a full-range (38Hz-22.5kHz ±3dB), baffleless, time- and phase- mica-filled polymer skin. The 4.5-inch transmission-line-loaded accurate speaker system with an 8-inch cast frame, curvilinear midrange driver is the patented Vandersteen open frame polycone woofer and a 1-inch alloy dome tweeter. Model 1Cs unit that has no reflective magnet structure behind the cone. sell for $785 per pair! That is not a misprint. Vandersteen 1Cs High frequencies are reproduced by a 1-inch ceramic coated cost seven hundred eighty-five dollars a pair. alloy dome, dual chamber tweeter. There is a rear facing .75-inch alloy dome auxiliary tweeter and a switch to turn it off. The Model 2Ce Signature is a baffleless, time- and phase- Vandersteen Model 5s cost $10,900 a pair. accurate 4-way (Vandersteen calls it a 3-way) speaker system with a 10-inch active bass coupler (low-woofer), an 8-inch cast frame polymer cone woofer, a 4.5-inch linear surround mid- Drivers range (same cone assembly as the Model 3 but mounted in a The drivers used in the Vandersteen speaker systems range conventional die-cast frame with ceramic magnet), and a 1-inch from near-catalog units, sourced from Vifa/Scanspeak and alloy dome tweeter. SEAS, to unique Vandersteen designs protected by patents. All drivers have Vandersteen part numbers and are made to pro- Each Model 2Ce Signature is hand-tweaked in an anechoic prietary specifications. chamber to produce response from 32Hz-21kHz within ±1.5dB. Pairs are matched to .1dB (one tenth dB) and sell for $1,549.
The Model 3A Signature is a baffleless, time- and phase- accurate 4-way (Vandersteen calls it a 3-way) speaker system with a 10-inch cast frame aluminum cone active bass coupler with dual spider suspension, an 8-inch cast frame polymer cone woofer, a 4.5-inch patented “open-frame” midrange and a 1-inch dual-chamber alloy dome tweeter (using Model 5 tech- nology). All drivers are assembled overseas from parts obtained from Each Model 3A Signature is hand-tweaked in an anechoic sources around the world. Vandersteen owns the tooling for all chamber to produce response from 30Hz-22kHz within ±1.5dB proprietary components, some of which are manufactured in (-3dB@26Hz). Pairs are matched to .1dB and sell for $3,495 Hanford. including Sound Anchor rear braces. Vandersteen tweeters are modified in the Hanford factory and The Model 5 is the Vandersteen flagship. It has a built-in pow- the level of modification varies with speaker model. The tweet- ered subwoofer based on a unique 12-inch dual aluminum cone ers used in the Models 3 and 5 are quite different from those driver with an enormous symmetrical push-pull motor system. used in the Models 1 and 2 although they all look similar from the front. This subwoofer is blended with the upper frequency drivers utilizing a passive high-pass filter system, which is placed in All drive elements have die-cast alloy frames (except the active front of the main amplifier, which drives the speaker above bass coupler in the Model 2) and large ceramic magnets are
32 www.audioperfectionist.com Vandersteen utilized in all drivers except for the open frame midrange unit, time smear improves transient response and makes the speak- which has an Alnico magnet that is made in England. ers sound more open and less “boxy.” The aerodynamic frame and small diameter Alnico magnet used in this driver provide Vandersteen woofers have linear magnetic systems with an open path from the rear surface of the diaphragm to the extended and shaped pole pieces and copper rings. The transmission line of the enclosure, where the rear wave is ter- 10-inch active bass coupler used in the Model 3 has a filled minated without harmful reflections. aluminum cone and a dual spider suspension for extremely long linear travel. The chassis for the symmetrical push-pull Vandersteen uses metal alloys for subbass driver diaphragms, subwoofer driver used in the Model 5 is die-cast in Hanford. which, in his designs, do not radiate directly toward the listener. He prefers metal alloys for tweeters, too. The metal domes in The “open frame” midrange driver used in the Models 3 and 5 Vandersteen tweeters are mechanically damped to minimize (patent number 5073948) has a curvilinear mica-filled polymer the inevitable resonant peak that occurs at ultrasonic frequen- cies.
Vandersteen has chosen a mica-filled polymer material for the diaphragms of his midrange drivers. For this critical range of frequencies, he believes that this material provides the best compromise between softer compounds like paper and stiffer alternatives like metal alloy. We have already discussed, at length, the arguments about diaphragm materials in the articles profiling Dunlavy and Thiel. Vandersteen offers another balance of compromise for you to evaluate by listening.
Dunlavy uses soft diaphragm materials exclusively. Thiel uses primarily aluminum alloys. Vandersteen uses aluminum for sub-
Vandersteen symmetrical push-pull subwoofer driver bass, metal alloys for high frequencies and high-tech polymers for midrange diaphragms. cone and a flat, linear surround. The die-cast chassis for this driver is manufactured in Hanford, the magnet assembly comes Enclosures from England, the cone is made in Germany and the driver is Vandersteen uses individual subenclosures to create smaller assembled in Scandinavia. This very special midrange is partly baffles around each driver than those found in other speakers. responsible for the extremely open, spacious sound of the These structures are complex and commensurately difficult to Vandersteen Model 3 and Model 5 speakers. Here’s why. make. Computer controlled woodworking machinery is required to manufacture the elaborate shapes of the various parts used to create the Vandersteen enclosures. Conventional midrange drivers have large magnet assemblies directly behind the radiating diaphragm (cone or dome). Sound Vandersteen has designed and made specialized machinery coming from the back of the diaphragm is reflected off the to automate the manufacturing process but much work must magnet and frame of the driver and is reradiated through the still be completed by hand. For example, there are 32 pieces diaphragm with a slight delay. Planar speakers don’t suffer from of MDF material, in different sizes and shapes, used to make this problem and that’s one reason that planar speakers tend to each Model 2 speaker. These parts are created by computer sound more open than conventional dynamic speakers. controlled saws and multiaxis routers and assembled using custom-built pneumatic jigs and fixtures. The Vandersteen open frame midrange driver eliminates these reflections as a source of time smear. Removing this reflective Models 1, 2, and 3 have enclosures made from heavily braced
www.audioperfectionist.com 33 Vandersteen
MDF material. Panels in the Model 3 are 1-inch to 1.5-inches between drivers. All components are of exceptionally high qual- thick. The trapezoidal shape of the bass enclosures adds stiff- ity, including the unique barrier strip input connectors which ness to the structure and helps to break up internal standing facilitate biwiring. waves. All glass-epoxy circuit boards are double-sided and through- plated. Premium resistors, premium film capacitors and air-core Cabinet resonances have been specifically engineered to be inductors wound with 6n copper wire are used throughout. dispersive rather than additive by shaping and arranging the many small parts used to form the enclosures. The mechani- Critical components are encapsulated in potting resin to make cal design of the Vandersteen cabinets was accomplished with them immune to vibration. The crossover networks in the Model the aid of FFT analysis. An accelerometer was placed on all 5 use battery-biased film capacitors as do the external high- cabinet surfaces and the output analyzed with a computerized pass filters used with the Model 5 speaker system. instrument during the design process. Models 1, 2, & 3 have unique thermal protection circuits which prevent damage to the midrange and tweeter drivers in the event of a severe overload. This makes these models ideal for home theater use.
Quality Control Vandersteen minimum baffle enclosure design All speaker components are individually performance tested. Drivers are selected, matched and coded for response char- The subwoofer enclosure for the Model 5 is shaped like a acteristics. Driver response characteristics are recorded to truncated pyramid and made from constrained-layer damped, aid in speaker matching and to facilitate future replacement. 2-inch thick laminated MDF sitting on a base made from 1-inch Crossover components are tested and selected before assem- thick machined epoxy laminate material. This bass enclosure bly. Completed crossover networks are individually tested and has a series of unique braces throughout its interior (see pic- matched to .1dB tolerances. tures on next page and in Audio Perfectionist Journal #12). The head module for the Model 5 is made from 22 laminated layers of .75-inch MDF. The front plate of the head module, All speakers are high-power sweep tested and performance on which the drivers are mounted, is made from a 1-inch thick verified by computer instruments. All “Signature” models, and machined epoxy laminate material. the Model 5, are individually tested in an anechoic chamber where crossover components are hand tweaked to achieve The Models 1, 2, & 3 have hardwood-veneered end caps, the advertised performance specifications. Speaker pairs are available in a variety of finishes, at the top and bottom of cloth- matched to .1dB (one tenth dB). wrapped enclosures. Model 5 cabinets are hand finished and inspected at all stages The Model 5 has a furniture-grade wood finish on the lower of assembly. portion of the cabinet with a cloth-covered bonnet crowning the upper portion. Model 5s are available in a variety of standard and custom wood finishes. Strong Points Vandersteens offer a unique combination of smooth, musical sound and high resolution. When driven by the best electronic Crossover Networks components, Vandersteen speakers produce a more focused The crossover networks in all Vandersteen speakers are engi- and three-dimensional image than most other speakers, regard- neered to provide frequency and phase compensation for the less of type or cost. You hear more of the signal and less of the drive elements and to provide first-order acoustic transitions speaker, in my opinion.
34 www.audioperfectionist.com Vandersteen
They cost far less than speakers with fancier cabinets and Vandersteen speakers, with the exception of the Model 5, comparable component quality. Each model in the Vandersteen may seem to be dynamically compressed to some listeners. I line provides performance that can be favorably compared to believe that this is due to a reduction in distortion, particularly speakers costing twice as much or more. baffle distortion, to which many listeners have become accus- Models 2 and above can be biwired or passively biamped. tomed. Subwoofers can increase the perceived and actual Vandersteen subwoofers blend seamlessly with all models dynamic range potential of most Vandersteen models. except the 5, which has a built-in powered subwoofer. Even the lower-priced models in the line provide full-range response with bass extension that is unrivaled by competitors The low price tags on Vandersteen speakers may be confusing costing far more. to the novice buyer. These are high resolution loudspeakers that should be used with associated equipment of the highest quality. They are seldom demonstrated or reviewed that way. Consumers and reviewers who have been told that they should spend two-thirds of their audio budget on speakers will need to be reeducated.
Measurements Vandersteen speakers deliver impeccable measured perfor- mance. They are demonstrably accurate in both frequency and time domains. The alloy dome tweeters produce some minor ringing on the impulse and step response graphs. This is due to energy storage at ultrasonic frequencies. Is this a reasonable Interior of Vandersteen Model 5 subwoofer enclosure trade-off for higher resolution? We’ll discuss this question in the next Audio Perfectionist Journal. Negative Considerations Vandersteen speakers do not present a difficult load to the The Vandersteen measurements show a wider range of fre- amplifier but they do have low nominal impedance and they quencies, 100Hz to 40kHz, and have 5dB increments per divi- are relatively low in sensitivity. They will require more amplifier sion. These graphs are not accurate below 300Hz. Ignore the power than higher sensitivity designs and they will not play as lower portion of the range between 100 and 300Hz. Note that loud as speakers with higher sensitivity or steep-slope cross- the tweeter resonance is well above the audible range and that over filters. high frequency bandwidth extends far above the range of soft dome tweeters to -6dB at 30kHz. APJ
While inexpensive amplifiers will drive Vandersteens, the short- comings of less than state-of-the-art electronic components will be mercilessly presented to the listener by these high resolution speakers.
The unique bass loading system of the Vandersteen Models 2 and 3 allows these products to produce lots of bass energy down to very low frequencies. Some listeners may prefer a tighter, leaner low end with more definition. Vandersteen sub- woofers provide the solution here. Vandersteen Model 5 head module
www.audioperfectionist.com 35 Vandersteen
Vandersteen 2Ce
36 www.audioperfectionist.com Vandersteen
Vandersteen 2Ce
www.audioperfectionist.com 37 Vandersteen
Vandersteen 3A Signature
38 www.audioperfectionist.com Vandersteen
Vandersteen 3A Signature
www.audioperfectionist.com 39 Journey to Enlightenment
Journey to Enlightenment but I wasn’t nearly as involved with it emotionally. At the time I by Shane Buettner thought this was simply because I was watching more movies and reading more books. But, as I found out, it was something From bipolar speakers to time- and phase-accuracy: a more. description of my personal journey into high-end audio. Six or seven years back, I swapped out the Harman Kardon As a complement to Dick Hardesty’s work on time- and phase- integrated amp that I had used through college for a Pro Logic accurate speakers, Dick thought it would be interesting for surround sound A/V receiver. I had a big screen TV and a you to read a testimonial regarding my own experiences in laserdisc player, and I enjoyed watching movies on this system the last year. This time period started with me using Definitive very much. But I found I was listening to music more often in Technology bipolar loudspeakers and, after experiencing sev- my car than at home. Then one day I upgraded my laserdisc eral high-end speaker systems, culminated with me buying a player from a lower end Pioneer player to a Pioneer Elite reference system of Vandersteen loudspeakers (just as Dick CLD-79. This not only improved the video end of the movie predicted I would after gaining more experience). experience, I accidentally rediscovered music as an emotional experience through the CLD-79, which was a decent CD player. For me, this journey is very much about being exposed to a higher standard and set of perspectives on audio than the cur- rent home theater gurus promulgate. The attitude that products It’s good enough for out there are “good enough for home theater” is keeping people from being exposed to truly high-end audio products in many home theater... cases. This is pervasive in the industry, from what you’ll read in the magazines that support the industry, to what you’ll likely be All of a sudden, instead of a congealed tangle of sounds cre- shown on a showroom floor by dealers of home theater prod- ating each song, I could hear the instruments that made up ucts. But trust me, there is a higher fidelity, truer experience the band when the music played. All of a sudden digital didn’t to be had if you’re willing. I wrote a song about it, and now I’m sound as bad, and I began to focus much more on music and gonna’ play it for ya’…. became emotionally involved with it once again.
From that point on music playback began to enter into my pur- My Roots chasing decisions again, but getting access to reliable informa- I’m 31 years old. As long as I can remember I’ve been emotion- tion on which products fit that particular bill became another ally involved with books, music and movies, and not necessarily obstacle. As much as I’d like to debunk some of the myths in that order at any given time. I was an avid vinyl collector in regarding front-end electronics and their suitability for music as my later years of high school, and throughout college. well as home theater, I need to stick to speakers, so, now that you have some background on me, let me refocus. A college roommate and I collected records and shared a sys- tem to play them on. (At one point we even had a huge pair of Acoustat electrostatic speakers that we used to cram into the The Home Theater Years tiny apartments we lived in!) When we went our separate ways Over the last several years I’ve used two brands of loudspeak- the turntable went with him and, with the CD boom in full force, ers for the most part: NHT and Definitive Technology. I bought I started collecting CDs from that point on. It was the most con- an NHT system several years ago based on their home theater venient thing to do. model VT-2. NHT builds a good and honest loudspeaker that’s well engineered, and provides a hell of a lot of performance for CD collecting was coincident with an increased involvement the money compared to many conventional loudspeaker manu- with film and movies on home video. I bought a laserdisc player facturers. I used the VT-2 system for two to three years before to experience films in their original aspect ratios and remained swapping it out for another NHT system, this one based on the very focused on movie watching. I still listened to a lot of music 2.5i model, which I found to be more satisfying musically.
40 www.audioperfectionist.com Journey to Enlightenment
Over a year ago I swapped out the NHT 2.5i-based speaker the bipoles I had been listening to. But hearing this effect as it system for a Definitive Technology system based on a pair of actually is on the disc was maddening. their “powered tower” bipolar loudspeakers with built-in pow- ered subwoofers, which I had reviewed for Widescreen Review. Instead of Lovett being a palpable vocal presence in the front of the room, which is a reasonable impression of Lyle being in the The Definitives are also a relatively good buy for the money, room on the regular CD, the multichannel mix on the DTS disc particularly for home theater. They’re very dynamic, to say the created a big, fat, lifeless image emanating from everywhere least, with the home theater slam and impact that many people but nowhere in particular. And that’s to say nothing of the rest enjoy. The resolution overall is pretty good, about as good as of the mix. You can argue about whether you want to hear how the NHTs, but more laid back compared to the rather forward ludicrous some of these mixes are, which accurate speakers and matter-of-fact sound of the NHTs. will certainly show you.
Of course, the signature of the Definitives is the bipolar radia- tion pattern created by having driver arrays mirrored on the It was like going front and back of the speakers, firing in phase with one anoth- er. The seductive thing about them is that, in a room like mine, to the eye doctor and which is a very typical tract home living room doubling as a home theater/listening room, they add a sense of spaciousness getting a new prescription. without the other colorations and dynamic limitations that elec- trostatic (or hybrid electrostatic) designs offer. Up at the front of the soundstage, with two-channel material, the differences between the Definitive and Aerial systems were The soundstage of the Definitives sounds relatively focused just as obvious. Instead of an artificially large soundstage up compared to other conventional designs, but is artificially front, the Aerials offered a sharp perspective with space and expanded by the reflected radiation coming off the front wall. air around vocals and instruments, and sound that was not at What I didn’t notice until living with the Definitives for several all confined to the physical boundaries of the speakers. But the months, followed by living with a superior monopole speaker Aerial’s spaciousness did not come at the expense of detail system by Aerial Acoustics (that I also reviewed for Widescreen and resolution, which was inevitably the case with the bipoles, Review), is that this artificial spaciousness is entirely detrimen- whose soundstage was undoubtedly blurred as a result of all tal to image focus and was not allowing me to truly hear what the reflected energy bouncing off the front wall from the rear- was happening in the soundstage of music and to the spatial firing driver arrays. characteristics of multichannel music and movies. The Aerials also did a better job of sounding less like big When the Aerial Acoustics speaker system came in, the speakers with big cabinets. The Aerials are built to a higher improvements were immediate and dramatic. The midrange, in standard, with superior damping and extensive cross-bracing. particular, came to life, imparting a dramatic sense of increased This results in the Aerials doing a much better job of keeping resolution. Spatially, sonic images became very sharply the cabinet from singing along with the music. The midrange focused; it was like going to the eye doctor and getting a new and tweeter of the Aerials are isolated in separate subenclo- prescription. sures, with the midrange surrounded by asphalt and wool to prevent coloration in that critical frequency range. This con- With some of the multichannel music material that I had previ- struction quality allowed the Aerials to “disappear” in the room ously appreciated, the increased focus revealed spatial and to an extent that the Definitives really couldn’t approach. spectral clues that I simply hadn’t heard before. On the DTS CD of Lyle Lovett’s Joshua Judges Ruth I was able to hear that The experience I’ve gained in the year since I wrote that review the mixes on several of the songs put Lovett’s voice into all has been invaluable, and the things I look for in a loudspeaker five channels creating a vocal image right inside my head. This don’t allow for the colorations I hear now in bipolar loudspeak- effect had actually been mitigated by the poor image focus of ers. I like and respect the people I know at Definitive and I
www.audioperfectionist.com 41 Journey to Enlightenment hope this isn’t taken to sound as critical as it might. I do believe At Dick’s place, when he played a reference quality LP like Mel that Definitive offers products superior to many things out there Torme and Friends Live at Marty’s, it was spooky in its “you at the same price and even higher. But today my standard is are there” feeling that particular recording imparts. The sound elevated, and let me tell you why. of the crowd, the musicians on stage, the dual vocalists coming from different sides of the stage—it was all there so clearly that I could practically see it. That’s imaging! Exposure to Time- and Phase-Accurate Speakers During the months that I had the Aerial Acoustics speaker sys- tem in my home, I also spent a lot of time listening to music And I could certainly feel it; the connection to the music was and movies on Dick’s Vandersteen system. There were several physical, just like being there. Needless to say, this experience things I noticed immediately about the Vandersteen system that also renewed my interest in vinyl, but that’s a different story for I would not hear from anything else I listened to. a different day!
The first thing that struck me was how open and transparent the sound from the Vandersteens was in general, and the mid- This sensation of the range in particular. Relaxed, liquid, musical detail and low level resolution that just poured out of… well, nothing. Vandersteen’s open sound...
The “baffleless” design of the Vandersteens means that there is is interesting to contrast with no baffle around the midrange and tweeter module at all. I think the main thing that people are seduced by with electrostats is Dunlavy’s speakers. that you don’t hear a box around them; they just seem to disap- pear in the room. The Vandersteen baffleless design imparts Moving on, Dick also played some cuts off of the Q-Sound much of that same sense of openness and lack of cabinet col- recording of Roger Waters’ Amused to Death, and some oration. Smooth, grain-free sound with lots of air and extension multichannel music such as Columbia’s DVD of the Best of is what I heard from the Vandersteens. Sessions at West 54th Volume 1. With the Roger Waters disc, which uses a bunch of phase-related shenanigans to create This sensation of the Vandersteen’s open sound, uncolored by some very interesting spatial effects, the Vandersteens placed large cabinet structures, is interesting to contrast with Dunlavy’s sounds so far out to the sidewalls that, if I closed my eyes, it speakers, another time/phase-accurate, well-engineered sounded like I was in a much larger room, an absolutely enor- design. While there were many things I admired listening to mous space. Sounds also imaged directly above my head at Dunlavy speakers, I was always aware of those speakers being the listening position, an amazing sensation I’d not experienced just what they are: very big speakers with enormous cabinets. with any other loudspeaker. Although we were listening in stereo, to the uninitiated I think it might have seemed that the At the CES show in January 2001 Dunlavy’s room was set up Vandersteen VSM surround speakers on the walls were active with several of his speaker pairs lined up next to one another. (they weren’t), so expansive was the imaging. It was always easy to discern exactly which speaker pairs were playing at any given time. The Dunlavy’s are certainly more The Sessions DVD features a Dolby Digital 5.1-channel track of efficient and will play louder and perhaps more dynamically a mix that’s not nearly as offensive as many of its multichannel than the Vandersteens, which will be a strong recommendation brethren. It’s an interesting sonic perspective that moves with for many people. But I just didn’t get the same sense of open the camera around the musicians, but it’s subtly and tastefully transparency that the Vandersteens offer in spades, and in the done. The performances are tremendous. end I found I couldn’t live without that. This disc illustrates how precise the Vandersteens are spatially, The other thing that knocked me out with the Vandersteens was especially on the Rickie Lee Jones cut, which has sound mov- the precise and defined soundstage and unparalleled imaging. ing along the sidewalls and in between the front and surround
42 www.audioperfectionist.com Journey to Enlightenment speakers in a fairly aggressive but still natural fashion. My system is identical to Dick’s with 3A Signatures in the front, Suzanne Vega’s Caramel and Annie DiFranco’s 32 Flavors a VCC-5 center speaker, wall-mount VSM surrounds, and four are other examples, with everything from backing vocals and 2WQ subwoofers for the right and left front and right and left drums to accordions wrapping gently around the soundstage surround speakers, respectively. While this system isn’t going and into the surrounds as the camera pans. Neat stuff, and the to produce action movie explosions louder than they’d be in Vandersteens make it feel natural, real and emotionally involv- real life, the Vandersteens, with the 2WQ woofers at the four ing. primary channel positions, have enough transient snap and oomph that I’ve not missed any of the previous speakers I’ve Emotional involvement is a key issue here. As you’ll read in the had in my room when I watch movies. Journal, Dick’s belief is that time- and phase-accuracy are of such paramount importance that experienced high-end enthu- The imaging in between all speaker positions that made the siasts should only consider loudspeakers that are accurate in Sessions DVD so involving is also revelatory with movie mate- that respect—which means Thiel, Dunlavy, and Vandersteen rial. This system presents the most precise 360-degree imag- would be your only choices out there in the wide world of ing I’ve heard yet in my room. And what’s more, the time- and speakers. phase-accurate VSM surrounds do a better job of providing diffuse ambience (when called for) than any type of diffuse- After listening to the Aerial Acoustics speakers for a long radiating speaker I’ve heard. time—which are some of the best (if not the best) conventional speakers I’ve heard—I must agree with Dick that the time- and phase-accurate designs are in another performance category ...time- and phase-accurate altogether. Thiel, Dunlavy and Vandersteen all offer superb loudspeakers but, for the reasons outlined above, I connected designs are in another perfor- emotionally to what I heard from the Vandersteens. You may feel differently and connect emotionally to one of the other two mance category altogether... brands, which are indeed outstanding loudspeakers, too. This improvement is an order of magnitude above the perfor- mance you’d get from dipole surround speakers or DSP-based Reviewers can help point you in the right direction by accurate- decorrelation from a digital controller. Further, this speaker ly describing the performance attributes of a component, and system is evidence of the absurdity of the extended surround hopefully, in some cases, ferreting out the genuine products sound “features,” such as Surround EX and DTS-ES, now from the snake oil. Some people want us to do all the work, as being peddled by the manufacturers of digital controllers and we’re constantly bombarded with people who say “I’ve read all A/V receivers. the reviews you’ve done, but is product X better than product Y?” Do yourself a favor, use the reviews as a guide, and go out The spatial precision of the Vandersteen system is such that I and listen. Find out which of these recommended products you get dead-center phantom images between the surround chan- respond to emotionally and I think you’ll be happiest with your nels that are much more natural and “real” sounding than any purchases. hard center channel could be in the back of my room. And I know of what I speak.
At Home with the Vandersteens The Vandersteens had a few more tricks up their proverbial During the time that I had the Definitive Technology speakers sleeves that I discovered upon having them in my own home. in my room, I had enough of them that I tried several configu- I knew from listening at Dick’s that I was getting tremendous rations of extended surround: two speakers at the side of the two-channel performance with the Vandersteen system I pur- listening position with one or two speakers along the back wall, chased. But I wasn’t prepared for just how stellar they’d be with as well as two speakers on the back wall and a single center movie soundtracks. speaker in between.
www.audioperfectionist.com 43 Journey to Enlightenment
With the Definitives this brute-force approach yielded some This is to say nothing of the fact that most people have a hard increased performance with movie soundtracks encoded in time cramming 5.1-channels of speakers into their living rooms, extended surround (which should have been a clue regard- let alone 6.1, 7.1, or, God forbid, 10.2 channels! ing the imaging capability of the Definitive bipolars). But in my room, which measures about 13 feet across and about 18 feet long, the Aerials first, and then the Vandersteens especially, And Did I Mention the Price? sounded much better with any and all soundtracks in a straight- On top of the fact that I found the Vandersteens I purchased up 5.1-channel configuration due to the fact that these speak- offer superior performance to anything else I listened to, they ers image in a much more accurate fashion. cost little more than the Definitives and far less than other high- end competitors’ products. In all but the largest rooms I don’t believe that any more than 5.1-channels of accurate, time- and phase-correct speakers are The cost of my entire Vandersteen system is $11,619. This is necessary for realistic, high resolution surround sound. certainly not chump change, but look at everything that’s in this system: a pair of 3A Signatures ($3,495/pr, stands included), a VCC-5 center speaker ($1,995), a pair of VSM surrounds These are some of the most ($949/pr), and four 2WQ subwoofers ($1,295 each.) That’s a hell of a deal for nine loudspeakers, four of which have on- advanced and best performing board power amplifiers, and all of which contain proprietary components and technology. speakers available. The Aerial Acoustics speakers I reviewed, for example, had a Of course, all of this is contrary to what some people in the system cost of $25,800 with just two of their subwoofers, and industry are saying. There are some people saying that the $15,000 with no subwoofers. alleged success of Surround EX and DTS-ES proves that peo- ple want more channels. Some even cite decades-old studies The cost of a pair of Vandersteen 3A Signatures and two woof- that “prove” the more channels you have the better and more ers is $6,085. This combination offers time- and phase-accurate “realistic” the sound is. performance that is accurate in the frequency domain to toler- ances of ±3dB from 18Hz to 22kHz. Name another speaker, Tomlinson Holman himself trumps up a demo of a 10.2-channel time- and phase-accurate or otherwise, that offers the kind of system at every trade show, touting that system as the height performance and accuracy that the Vandersteen package does of realism. All I can say in response is that I seriously wonder at anywhere near this price. if any of these people have ever heard a properly set-up sys- tem of time- and phase-accurate speakers. If they had, I doubt These are some of the most advanced and best performing they’d be pushing systems of inferior products that require speakers available, and yet they’re also among the most rea- so many channels to create sound that is still grossly inferior sonably priced. What else is there to say? to that attained by my Vandersteen 5.1-channel system. Of course, I think this is a moot point anyway. That’s All Folks For more information about the speaker manufacturers profiled in this issue visit their web sites. Surround EX and DTS-ES may have helped some manufac- turers sell some A/V receivers, but on the content side DVD Dunlavy Audio Labs: http://www.dunlavyaudio.com producers haven’t even bothered to indicate on DVD packaging when a soundtrack is presented in one of these formats. And Thiel Audio: http://www.thielaudio.com my guess is that they haven’t been flooded by complaints from consumers who care. Vandersteen Audio: http://www.vandersteen.com
APJ
44 www.audioperfectionist.com Conclusion
Paula T. Hardesty, Publisher Richard L. Hardesty, Editor Edith Hardesty, Copy Editor Shane Beuttner, Equipment Review Editor Rick Johnson, Art Director
Conclusion Flash Forward to 2008 by Richard Hardesty Since this Journal was written in 2001 there have been several significant changes. While the industry continually changes This Journal included only brief comments about the sound concepts and objective measurements remain the same. quality of the products described. We’ll discuss the sound of Dunlavy Audio is no longer in business. Thiel and Vandersteen Dunlavy, Thiel and Vandersteen speakers in greater detail in have improved their products and introduced new models. the Journal #8. We’ll talk more about time- and phase-accura- Meadowlark Audio assumed the position of the third major cy and its importance in stereo systems and how it affects the manufacturer to offer a complete line of time- and phase-accu- performance of home theater systems, too. rate loudspeakers and subsequently went out of business.
I’ll tell you the truth about center channel speakers and Shane Buettner became a reviewer for The Absolute Sound describe why most of them perform so poorly and we’ll talk magazine and Video Technical Editor for The Perfect Vision about the surround channels in a multichannel system, and I’ll magazine, and Equpment Review editor for the Audio describe some special requirements for accurate reproduction Perfectionist Journal. He then became editor of Ultimate AV from speakers positioned at the sides or back of the room. APJ on the web and now is the editor of Home Theater magazine.
Audio Perfectionist Journals #9-16 offer a book-length refer- ence on home audio systems. They’re available from http://www.audioperfectionist.com APJ
17141 Los Robles Circle, Fountain Valley, CA. 92708
www.audioperfectionist.com 45